Full-view optical remote sensing image simulation method

Abstract

The invention discloses a full-view optical remote sensing image simulation method. The method comprises the following steps: obtaining an imaging position of a satellite at a specified moment according to a satellite orbit dynamic model; obtaining the direction of the camera in the inertial coordinate system; obtaining the imaging vector direction of each pixel on the camera detector according to the direction of the camera in the inertial coordinate system; obtaining a camera imaging geographic position according to an intersection point of the imaging vector direction of each pixel on the camera detector and a preset three-dimensional scene model base map; the radiation intensity of each imaging point is obtained according to the camera imaging geographic position, and the pixel value of each imaging point is obtained according to the radiation intensity of each imaging point, the detector noise and the quantum efficiency. The method overcomes the defect that an existing remote sensing image simulation method cannot generate inclination angle remote sensing image simulation, and can simulate and generate a high-precision remote sensing image with any inclination angle.

Description

Technical Field

[0001] This invention belongs to the field of optical remote sensing image simulation technology, and particularly relates to a full-view optical remote sensing image simulation method. Background Technology

[0002] Optical remote sensing image simulation simulates the physical processes of optical remote sensing imaging, mimicking the generation of remote sensing images to predict the effects of optical remote sensing imaging. Early optical remote sensing simulations primarily used physical or semi-physical simulations. However, with the rapid development of computer technology, digital simulations now mainly utilize computer graphics, numerical computation, and other computer technologies.

[0003] Currently, widely used and relatively comprehensive foreign commercial simulation tools include the Vega series simulation modules from the United States and SE-workbench from France. The former mainly consists of modules such as Vega, SensorVision, and SensorWorks, and has the ability to simulate complex 3D scenes; while the latter can simulate the radiation characteristics of 3D scenes, and can simulate high-precision remote sensing images after setting appropriate sensor parameters. Domestically, ORSIS optical remote sensing image simulation software uses atmospheric radiative transfer software and atmospheric characteristic databases to simulate optical remote sensing images. However, the above-mentioned remote sensing simulation methods or systems mainly focus on simulating radiation information during remote sensing imaging, that is, the realism of the image, and give less consideration to the geometric quality of the image, including geometric resolution, geometric deformation, and geometric position.

[0004] With satellite remote sensing entering the high-resolution era, on-orbit optical remote sensing images have achieved sub-meter resolution, and image positioning errors have improved from hundreds of meters to tens of meters. Optical satellite imaging modes have become more flexible and diverse, evolving from traditional nadir point imaging to large-angle tilt imaging in any direction. These advancements have placed many new demands on optical remote sensing simulation technology, including the need to consider the effects of on-orbit deformation and vibration of the satellite platform, the influence of occlusion relationships between surface features, and the realistic simulation of the sides and shadow areas of features. Traditional general-purpose remote sensing image simulation systems struggle to meet these requirements in terms of imaging geometry simulation accuracy and platform influence simulation. Dedicated remote sensing image simulation systems that use high-resolution two-dimensional images as simulation base maps lack the ability to simulate the sides of features and cannot effectively simulate the occlusion relationships between actual features, thus failing to meet these requirements. Summary of the Invention

[0005] The technical problem solved by this invention is to overcome the shortcomings of the prior art and provide a full-view optical remote sensing image simulation method. This method overcomes the limitation of existing remote sensing image simulation methods in being unable to generate remote sensing image simulations at tilt angles, and can simulate and generate high-precision remote sensing images at arbitrary tilt angles.

[0006] The objective of this invention is achieved through the following technical solution: a full-view optical remote sensing image simulation method, comprising: obtaining the imaging position of the satellite at a specified time based on a satellite orbital dynamics model; obtaining the orbital coordinate system pointing of the satellite orbital position at the specified time based on the satellite orbital dynamics model; obtaining the satellite attitude matrix based on the satellite attitude information; obtaining the satellite attitude pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite orbital position at the specified time and the satellite attitude matrix; obtaining the camera pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite orbital position at the specified time, the satellite attitude matrix, the imaging camera installation matrix, the installation attitude change matrix, and the camera attitude pointing change caused by micro-vibration; obtaining the imaging vector pointing of each pixel on the camera detector based on the camera pointing in the inertial coordinate system; obtaining the camera imaging geographical location based on the intersection of the imaging vector pointing of each pixel on the camera detector and a preset three-dimensional scene model base map; obtaining the radiation intensity of each imaging point based on the camera imaging geographical location; and obtaining the pixel value of each imaging point based on the radiation intensity, detector noise, and quantum efficiency of each imaging point.

[0007] In the above-mentioned full-view optical remote sensing image simulation method, the satellite orbital dynamics model is obtained through the following formula:

[0008]

[0009] in, The acceleration of the satellite in the J2000; The acceleration due to gravity at the Earth's center; This is the acceleration due to the non-spherical gravitational effect of the Earth.

[0010] In the above-mentioned full-view optical remote sensing image simulation method, the satellite attitude matrix R att It can be obtained through the following formula:

[0011]

[0012] Among them, R att For the satellite attitude matrix, Let ω be the Euler angle of the satellite attitude about the X-axis, ω be the Euler angle of the satellite attitude about the Y-axis, and κ be the Euler angle of the satellite attitude about the Z-axis.

[0013] In the above-mentioned full-view optical remote sensing image simulation method, the satellite attitude in the inertial coordinate system is oriented as C. att It can be obtained through the following formula:

[0014] C att =C orb *R att ;

[0015] Among them, C attC represents the orientation of the satellite in the inertial coordinate system. orb R is the orbital coordinate system pointing to the satellite's orbital position at a specified time. att This is the satellite attitude matrix.

[0016] In the above-mentioned full-view optical remote sensing image simulation method, the installation attitude change matrix R temp It can be obtained through the following formula:

[0017]

[0018] Wherein, δα1 is the change of Euler angles about the X-axis due to the change of camera mounting attitude on track, δβ1 is the change of Euler angles about the Y-axis due to the change of camera mounting attitude on track, and δγ1 is the change of Euler angles about the Z-axis due to the change of camera mounting attitude on track.

[0019] In the above-mentioned full-view optical remote sensing image simulation method, the camera attitude pointing change R caused by micro-vibration vib It can be obtained through the following formula:

[0020]

[0021] Wherein, δα2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the X-axis, δβ2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the Y-axis, and δγ2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the Z-axis.

[0022] In the above-mentioned full-view optical remote sensing image simulation method, the camera's orientation C in the inertial coordinate system is... cam3 It can be obtained through the following formula:

[0023] C cam3 =C orb *R att *R cam2body *R temp *R vib ;

[0024] Among them, C orb R is the orbital coordinate system pointing to the satellite's orbital position at a specified time. att R is the satellite attitude matrix. temp To install the attitude change matrix, R vib R represents the camera's attitude pointing change caused by micro-vibrations. cam2body Install a matrix for the imaging camera.

[0025] In the above-described full-view optical remote sensing image simulation method, the imaging vector pointing of each pixel on the camera detector is obtained by the following formula:

[0026]

[0027] Among them, C cam3 Here, X represents the camera's orientation in the inertial coordinate system, Y represents the pixel's imaging position in the inertial coordinate system, and Z represents the pixel's imaging position in the inertial coordinate system. s Let X and Y be the imaging position of the camera center pixel in the inertial coordinate system. s Let Y and Z be the Y-coordinates of the imaging position of the camera's center pixel in the inertial coordinate system. s Let X be the Z-coordinate of the imaging position of the camera center pixel in the inertial coordinate system, k be a constant coefficient, x′ be the X-axis component of the imaging vector of the pixel in the camera coordinate system, y′ be the Y-axis component of the imaging vector of the pixel in the camera coordinate system, and z′ be the Z-axis component of the imaging vector of the pixel in the camera coordinate system. d Let X be the component of the imaging vector of a pixel in the inertial coordinate system in the X direction, and Y be the component of the imaging vector in the inertial coordinate system in the X direction. d Let Z be the component of the imaging vector of a pixel in the inertial coordinate system in the Y direction. d The component of the imaging vector of a pixel in the inertial coordinate system in the Z direction.

[0028] In the above-described full-view optical remote sensing image simulation method, the geographical location of the camera image is obtained by the following formula:

[0029]

[0030] Where X1 is the X-axis coordinate of the camera's image geographic location, Y1 is the Y-axis coordinate of the camera's image geographic location, Z1 is the Z-axis coordinate of the camera's image geographic location, and X... s Let X and Y be the imaging position of the camera center pixel in the inertial coordinate system. s Let Y and Z be the Y-coordinates of the imaging position of the camera's center pixel in the inertial coordinate system. s Let X be the Z-coordinate of the camera center pixel's imaging position in the inertial coordinate system, k0 be the second coefficient, and X be the Z-coordinate of the imaging position in the inertial coordinate system. d Let X be the component of the imaging vector of a pixel in the inertial coordinate system in the X direction, and Y be the component of the imaging vector in the inertial coordinate system in the X direction. d Let Z be the component of the imaging vector of a pixel in the inertial coordinate system in the Y direction. d The component of the imaging vector of a pixel in the inertial coordinate system in the Z direction.

[0031] In the above full-view optical remote sensing image simulation method, the radiation intensity of each imaging point is obtained by the following formula:

[0032] f(x,y)=∑ u ∑ v L(xu,yv)h(u,v);

[0033] Where f(x,y) is the radiation intensity at the imaging point (x,y), x is the x-axis coordinate of the imaging point, y is the y-axis coordinate of the imaging point, L(xu,yv) is the radiation intensity of the image point within the region that contributes to the radiation energy of the imaging point, h(u,v) is the normalized point spread function, u is the x-axis range of the region that contributes to the radiation energy of the imaging point, and v is the y-axis range of the region that contributes to the radiation energy of the imaging point.

[0034] A full-view optical remote sensing image simulation system includes: a first module for obtaining the satellite's imaging position at a specified time based on a satellite orbital dynamics model; a second module for obtaining the orbital coordinate system pointing of the satellite's orbital position at a specified time based on the satellite orbital dynamics model, obtaining the satellite attitude matrix based on satellite attitude information, obtaining the satellite attitude pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite's orbital position at a specified time and the satellite attitude matrix, and obtaining the camera's pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite's orbital position at a specified time, the satellite attitude matrix, the imaging camera mounting matrix, the mounting attitude change matrix, and the camera attitude pointing change caused by micro-vibration; a third module for obtaining the imaging vector pointing of each pixel on the camera detector based on the camera's pointing in the inertial coordinate system; a fourth module for obtaining the camera's imaging geographical location based on the intersection of the imaging vector pointing of each pixel on the camera detector and a preset three-dimensional scene model base map; and a fifth module for obtaining the radiation intensity of each imaging point based on the camera's imaging geographical location, and obtaining the pixel value of each imaging point based on the radiation intensity, detector noise, and quantum efficiency of each imaging point.

[0035] Compared with the prior art, the present invention has the following advantages:

[0036] (1) Based on the actual physical process of remote sensing image acquisition, this invention models the entire process of each link and its error characteristics that affect the geometric accuracy of remote sensing images, such as orbit positioning, attitude determination, camera imaging, and atmospheric transmission, and truly considers the impact of each link of the imaging platform and remote sensor on the quality of the simulated image.

[0037] (2) The present invention uses a real scene 3D model as a simulation data source, which can accurately reflect the terrain undulation and the 3D shape of the surface features of the scene. In particular, it can accurately describe the facade information of surface protrusions such as buildings and realistically simulate the mutual occlusion of surface protrusions such as buildings in tilt angle imaging.

[0038] (3) The present invention adopts the radiation simulation method of ray tracing, which comprehensively considers the influence of the main environmental light sources such as solar radiation and ground object reflection on the simulation image, effectively solves the problem of image simulation in the shadow area, and realizes high-fidelity imaging simulation of ground objects. Attached Figure Description

[0039] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0040] Figure 1 This is a flowchart of the full-view optical remote sensing image simulation method provided in the embodiments of the present invention;

[0041] Figure 2 This is a schematic diagram of the orbital coordinate system provided in an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram of optical remote sensing imaging provided in an embodiment of the present invention;

[0043] Figure 4(a) is a schematic diagram of the point cloud representation of the three-dimensional model provided in the embodiment of the present invention;

[0044] Figure 4(b) is a schematic diagram of the voxel representation of the three-dimensional model provided in the embodiment of the present invention;

[0045] Figure 4(c) is a schematic diagram of the triangular mesh representation of the three-dimensional model provided in the embodiment of the present invention;

[0046] Figure 5 This is a schematic diagram of the point spread function provided in an embodiment of the present invention. Detailed Implementation

[0047] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0048] Figure 1 This is a flowchart of the full-view optical remote sensing image simulation method provided in an embodiment of the present invention. Figure 1 As shown, this full-view optical remote sensing image simulation method includes: obtaining the satellite's imaging position at a specified time based on the satellite orbital dynamics model; and obtaining the orbital coordinate system pointing to C where the satellite's orbital position at the specified time is located based on the satellite orbital dynamics model. orb The satellite attitude matrix R is obtained based on the satellite attitude information. att According to the orbital coordinate system pointing to C at the specified time, orband satellite attitude matrix R att Obtain the orientation C of the satellite in the inertial coordinate system. att According to the orbital coordinate system pointing to C at the specified time, orb Satellite attitude matrix R att Imaging camera mounting matrix R cam2body Install attitude change matrix R temp and camera attitude pointing change R caused by micro-vibration vib Obtain the camera's orientation C in the inertial coordinate system. cam3 According to the camera's orientation C in the inertial coordinate system cam3 The imaging vector pointing of each pixel on the camera detector is obtained; the geographical location of the camera imaging is obtained based on the intersection of the imaging vector pointing of each pixel on the camera detector and the preset 3D scene model base map; the radiation intensity of each imaging point is obtained based on the geographical location of the camera imaging; and the pixel value of each imaging point is obtained based on the radiation intensity, detector noise and quantum efficiency of each imaging point.

[0049] The satellite orbital dynamics model is obtained through the following formula:

[0050]

[0051] in, The acceleration of the satellite in the J2000; The acceleration due to gravity at the Earth's center; This is the acceleration due to the non-spherical gravitational effect of the Earth.

[0052] Satellite attitude matrix R att It can be obtained through the following formula:

[0053]

[0054] Among them, R att For the satellite attitude matrix, Let ω be the Euler angle of the satellite attitude about the X-axis, ω be the Euler angle of the satellite attitude about the Y-axis, and κ be the Euler angle of the satellite attitude about the Z-axis.

[0055] The orientation of the satellite in the inertial coordinate system, C att It can be obtained through the following formula:

[0056] C att =C orb *R att ;

[0057] Among them, C att C represents the orientation of the satellite in the inertial coordinate system. orb R is the orbital coordinate system pointing to the satellite's orbital position at a specified time.att This is the satellite attitude matrix.

[0058] Install attitude change matrix R temp It can be obtained through the following formula:

[0059]

[0060] Wherein, δα1 is the change of Euler angles about the X-axis due to the change of camera mounting attitude on track, δβ1 is the change of Euler angles about the Y-axis due to the change of camera mounting attitude on track, and δγ1 is the change of Euler angles about the Z-axis due to the change of camera mounting attitude on track.

[0061] Camera attitude pointing change R caused by micro-vibration vib It can be obtained through the following formula:

[0062]

[0063] Wherein, δα2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the X-axis, δβ2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the Y-axis, and δγ2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the Z-axis.

[0064] The camera's orientation C in the inertial coordinate system cam3 It can be obtained through the following formula:

[0065] C cam3 =C orb *R att *R cam2body *R temp *R vib ;

[0066] Among them, C orb R is the orbital coordinate system pointing to the satellite's orbital position at a specified time. att R is the satellite attitude matrix. temp To install the attitude change matrix, R vib R represents the camera's attitude pointing change caused by micro-vibrations. cam2body Install a matrix for the imaging camera.

[0067] The imaging vector pointing to each pixel on the camera detector is obtained by the following formula:

[0068]

[0069] Among them, C cam3 Here, X represents the camera's orientation in the inertial coordinate system, Y represents the pixel's imaging position in the inertial coordinate system, and Z represents the pixel's imaging position in the inertial coordinate system. sLet X and Y be the imaging position of the camera center pixel in the inertial coordinate system. s Let Y and Z be the Y-coordinates of the imaging position of the camera's center pixel in the inertial coordinate system. s Let X be the Z-coordinate of the imaging position of the camera center pixel in the inertial coordinate system, k be a constant coefficient, x′ be the X-axis component of the imaging vector of the pixel in the camera coordinate system, y′ be the Y-axis component of the imaging vector of the pixel in the camera coordinate system, and z′ be the Z-axis component of the imaging vector of the pixel in the camera coordinate system. d Let X be the component of the imaging vector of a pixel in the inertial coordinate system in the X direction, and Y be the component of the imaging vector in the inertial coordinate system in the X direction. d Let Z be the component of the imaging vector of a pixel in the inertial coordinate system in the Y direction. d The component of the imaging vector of a pixel in the inertial coordinate system in the Z direction.

[0070] The geographic location of the camera image is obtained using the following formula:

[0071]

[0072] Where X1 is the X-axis coordinate of the camera's image geographic location, Y1 is the Y-axis coordinate of the camera's image geographic location, Z1 is the Z-axis coordinate of the camera's image geographic location, and X... s Let X and Y be the imaging position of the camera center pixel in the inertial coordinate system. s Let Y and Z be the Y-coordinates of the imaging position of the camera's center pixel in the inertial coordinate system. s Let X be the Z-coordinate of the camera center pixel's imaging position in the inertial coordinate system, k0 be the second coefficient, and X be the Z-coordinate of the imaging position in the inertial coordinate system. d Let X be the component of the imaging vector of a pixel in the inertial coordinate system in the X direction, and Y be the component of the imaging vector in the inertial coordinate system in the X direction. d Let Z be the component of the imaging vector of a pixel in the inertial coordinate system in the Y direction. d The component of the imaging vector of a pixel in the inertial coordinate system in the Z direction.

[0073] The radiation intensity at each imaging point is obtained using the following formula:

[0074] f(x,y)=∑ u ∑ v L(xu,yv)h(u,v);

[0075] Where f(x,y) is the radiation intensity at the imaging point (x,y), x is the x-axis coordinate of the imaging point, y is the y-axis coordinate of the imaging point, L(xu,yv) is the radiation intensity of the image point within the region that contributes to the radiation energy of the imaging point, h(u,v) is the normalized point spread function, u is the x-axis range of the region that contributes to the radiation energy of the imaging point, and v is the y-axis range of the region that contributes to the radiation energy of the imaging point.

[0076] This embodiment mainly includes the following steps:

[0077] (1) Calculation of satellite imaging orbit position. Establish a satellite orbit dynamics model and calculate the satellite's imaging position at a specified time.

[0078] (2) Satellite imaging attitude calculation. Based on the satellite orbital dynamics model, the orbital coordinate system pointing of the satellite's orbital position at the imaging moment is calculated. Furthermore, satellite attitude data, imaging camera installation data, and satellite status data are used to accurately calculate the camera's attitude pointing.

[0079] (3) Camera imaging vector calculation. Using imaging parameters such as the camera principal point and principal distance, and further considering the effect of the atmosphere on the imaging light, the imaging vector pointing of each pixel on the camera detector is calculated.

[0080] (4) Calculation of camera imaging geolocation. A high-precision 3D scene model is used as the data base map. The intersection of the imaging vector and the 3D scene model base map is accurately calculated to obtain the accurate imaging position of the camera.

[0081] (5) Calculation of pixel values ​​for camera imaging points. By comprehensively utilizing parameters such as the ground reflection characteristics, environmental radiation characteristics, and camera radiation imaging characteristics of each imaging geographical location, the pixel value of each imaging point is calculated to obtain the simulation image.

[0082] Furthermore, the satellite's imaging orbit position is calculated. Satellites operating in Earth orbit follow orbital dynamics laws. Given the initial orbital values ​​and orbital dynamics equations at a specified time, the satellite's orbital position at various moments can be simulated. The orbital dynamics equations have the following form:

[0083]

[0084] In the formula, r is the position vector of the satellite in J2000; The acceleration due to gravity at the Earth's center; This is the acceleration due to the non-spherical gravitational effect of the Earth.

[0085] Furthermore, by calculating the satellite's imaging attitude and using the satellite orbital dynamics equations, the orbital coordinate system pointing to the C-axis at a given moment can be obtained. orb .

[0086] Based on the satellite attitude information at this time, the satellite attitude Euler angles can be used. This indicates that the satellite attitude matrix R is calculated. att ,

[0087]

[0088] Further, the pointing C of the satellite attitude in the inertial coordinate system is obtained. att ,

[0089] C att =C orb *R att

[0090] The imaging camera is mounted on a satellite platform, and its mounting matrix is ​​R. cam2body Then the pointing C of the camera in the inertial coordinate system can be calculated. cam ,

[0091] C cam =C orb *R att *R cam2body

[0092] During the camera's operation in orbit, its support structure undergoes structural deformation due to the alternating external heat flow environment, which also causes a shift in the camera's attitude orientation. This shift changes over time according to a certain pattern. Let the amplitude of the change in the camera's mounting attitude Euler angles be (δα1, δβ1, δγ1), and the resulting mounting attitude change matrix be R. temp ,

[0093]

[0094] Then, considering the camera's orientation after structural deformation,

[0095] C cam2 =C orb *R att *R cam2body *R temp

[0096] Some attitude control components on the satellite generate micro-vibrations of a certain frequency during operation, which can also affect the camera's attitude pointing. Let the equivalent attitude angle of the camera caused by the micro-vibration of the component be (δα2, δβ2, δγ2), then the change in camera attitude pointing caused by the micro-vibration is R. vib for

[0097]

[0098] After considering the influence of micro-vibrations, the camera attitude pointing is

[0099] C cam3 =C orb *R att *R cam2body *R temp *R vib

[0100] Furthermore, the direction of the camera imaging ray is calculated. The camera imaging ray consists of two parts: the starting position of the imaging ray and the imaging ray vector. According to the principles of optical imaging, the center of the imaging pixel, the optical imaging center, and the imaging area of ​​the ground object are on the same straight line. Therefore, the optical imaging center can be used as the starting position of the imaging ray. Based on the satellite orbit position calculated above and the camera imaging center's installation position on the satellite, the coordinate position of the camera imaging center [X] can be accurately calculated. s ,Y s Z s ].

[0101] Next, determine the direction of the imaging ray corresponding to each pixel. According to the geometric imaging model, let the camera focal length be f, the position of the camera principal point be (x0, y0) (in the camera coordinate system), and the position of the image point be (x, y). Then, the direction of the imaging ray corresponding to this image point is:

[0102]

[0103] Based on the camera attitude obtained in step two, the direction of the imaging ray in the Earth coordinate system can be further calculated. For imaging under certain pitch and roll angles, the refraction of light by the atmosphere can be further considered as needed to calculate the true direction of the ray after atmospheric refraction. The equation of the imaging ray can be expressed as:

[0104]

[0105] Furthermore, to effectively overcome the problems of traditional planar image base map data failing to represent surface protrusions such as buildings, resulting in positioning errors during tilted perspective simulations, and the inability to simulate building sides, this method uses a true 3D model that accurately represents surface undulations and feature shapes as the base map data. The intersection of the imaging ray and the base map is the true imaging geographical location, effectively avoiding the aforementioned problems or errors. The algorithm for calculating the intersection of the imaging ray and the 3D model is related to the representation method used for the surface 3D model. Common 3D model representation methods include point cloud representation, triangular mesh representation, and voxel representation. Taking triangular mesh representation as an example, calculating the intersection of the imaging ray and the triangular mesh model requires traversing all the constituent triangles in the triangular mesh model, determining the angle between the imaging ray and the normal of the imaging location area, and whether the intersection is obscured by other features. The intersection of the imaging ray and the triangular facet can be calculated using the following algorithm:

[0106] Let point P be a point on the triangular facet, and its coordinates be denoted as . The normal vector of the triangular facet is Then the parameters of the equation of the line corresponding to the intersection point If k0∈[0,+∞) and the intersection point is located inside the triangular facet, then the imaging ray intersects the triangular facet, and the coordinates of the intersection point are...

[0107]

[0108] Furthermore, to calculate the pixel value of the image point, the energy of the optical remote sensing image mainly comes from the radiation energy reflected along the imaging direction at the corresponding imaging position on the ground. It is necessary to calculate the area of ​​the ground imaging position corresponding to the image point, the bidirectional reflectance distribution function (BRDF) of the imaging position, and the radiation energy entering the imaging position area.

[0109] The radiation energy of optical remote sensing imaging mainly comes from the radiation energy reflected by the ground imaging location area along the imaging direction. The energy level is mainly related to the area of ​​the ground imaging location area, the two-way reflectance distribution function (BRDF) of the ground objects at the imaging location, and the radiation intensity entering the imaging location area.

[0110] The image point corresponds to the imaging location area. Let the detector pixel size be d, the camera system focal length be f, and the satellite imaging altitude be h. Then, the projection of the imaging location area corresponding to each image point on the imaging ray normal plane is approximately a square area with a side length of hd / f. Further calculate the angle θ between the normal of the imaging location area and the imaging ray. The area of ​​the imaging location area is approximately hd / (fcos(θ)).

[0111] The bidirectional reflectance distribution function of the imaging location area is related to factors such as ground feature type, light source direction, and observation direction, and can be selected from the model database or determined based on experience.

[0112] The radiant energy reflected from the imaging location mainly includes solar radiation and radiation reflected from surrounding ground features. The solar radiation intensity of the imaging location can be calculated using atmospheric radiative transfer models such as Modtran. For tilt angle simulations, since many imaging areas are located on the sides of ground features, the corresponding solar radiation intensity must be calculated based on the direction of solar radiation, the normal of the ground features, and the observation direction. The shading effect of surrounding surface protrusions must also be considered, i.e., determining the visibility of the imaging location to the sun. If the area is obscured by surrounding ground features and therefore invisible to the sun, this portion of solar radiation energy is not included. The intensity of radiation directly reflected by the sun is:

[0113]

[0114] In the formula The intensity of solar radiation reflected from the imaging location region. It is a two-way reflection distribution function. V represents the intensity of solar radiation received in the imaging location region. 0The visibility of the sun at this location is 1 if visible and 0 if invisible. S is the range of the imaging location area.

[0115] The reflection intensity of the ground object's reflected radiation is closely related to the three-dimensional shape of the imaging area and the reflectivity of the ground object, and cannot be directly calculated analytically. However, it can be solved using ray tracing. The contribution of radiation energy reflected from different directions to the imaging location area to the imaging radiation intensity is as follows:

[0116]

[0117] in Let be the bidirectional reflection distribution coefficient of the reflected light from direction Ω. Let be the reflected radiation flux from direction Ω.

[0118] Therefore, the total radiation intensity reflected from this imaging location is:

[0119]

[0120] In the above formula This represents the intensity of radiation reflected from a ground feature's reflection point, and its value can be solved recursively using the calculation method described above.

[0121] In optical remote sensing imaging, due to the diffraction effect of the camera's optical system, the image of a point light source on the ground is represented by an Airy disk distributed over a region on the image plane. Its energy distribution function is called the point spread function, and its shape is as follows: Figure 5 As shown. The diffraction effect of an optical system causes the energy reflected from the imaging location region of a single image point to radiate to surrounding pixels. Similarly, the image point also receives radiated energy reflected from the imaging location regions of surrounding pixels. Therefore, the radiated energy ultimately received by a certain image point is also affected by the radiated energy reflected from the surrounding region of its imaging location, and has the following form:

[0122]

[0123] In the formula, f(x,y) is the radiance at image point (x,y), L(xu,yv) is the radiance of image points within the region that contributes to the radiant energy of that image point, and h(u,v) is the normalized point spread function.

[0124] Based on the radiation intensity of a single image point, and taking into account characteristics such as detector noise and quantum efficiency, the DN value of that image point can be calculated.

[0125] The arbitrary-view optical remote sensing image simulation method provided in this embodiment mainly includes the following steps:

[0126] The first step is to calculate the satellite's orbital imaging position at a specified time based on the satellite's orbital dynamics model and the imaging simulation time. Satellites operating in Earth orbit follow orbital dynamics laws; given the initial orbital values ​​and orbital dynamics equations at a specified time, the satellite's orbital position at that time can be calculated. The orbital dynamics equations have the following form:

[0127]

[0128] In the formula, r is the position vector of the satellite in J2000; The acceleration due to gravity at the Earth's center; This is the acceleration due to the non-spherical gravitational effect of the Earth.

[0129] Earth's central gravitational acceleration It has the following forms:

[0130]

[0131] In the formula, M is the mass of the Earth and G is the gravitational constant.

[0132] Acceleration due to Earth's non-spherical gravitational effect It is related to the Earth's gravitational potential energy field U, which can be expressed as a spherical harmonic function series.

[0133]

[0134] Where R is the Earth's equatorial radius; (φ,λ) are the satellite's geocentric latitude and longitude in WGS84; and r is the distance from the satellite to the origin. For the normalized association Legendre function, These are the normalized harmonic function coefficients.

[0135] The acceleration due to Earth's non-spherical gravitational effect can be expressed as:

[0136]

[0137] Generally, based on the above equations, numerical integration methods are used to calculate the orbital position at each time point, such as the Runge-Kutta method and the Adams method.

[0138] The second step, based on the satellite orbital dynamics model, is to obtain the orbital coordinate system pointing to C where the satellite's orbital position is at a specified time. orb ,like Figure 2 As shown. Satellites typically adjust their imaging angle through attitude maneuvers to acquire satellite images from either a frontal or tilted perspective. Satellite attitude can be represented using Euler angles, quaternions, or orthogonal matrices; this implementation uses three Euler angles. This represents the satellite's attitude information; the satellite's attitude matrix R can be calculated using Euler angles.att :

[0139]

[0140] The orientation of the satellite in the inertial coordinate system, C att for:

[0141] C att =C orb *R att

[0142] The imaging camera is mounted on a satellite platform, given the camera's mounting matrix R on the platform. cam2body The pointing direction C of the camera in the inertial coordinate system can be calculated. cam for:

[0143] C cam =C orb *R att *R cam2body

[0144] During satellite operation, the camera attitude pointing is dynamically changed by various environmental factors, which in turn affects the imaging position and even image quality. For example, the camera support structure may deform due to the alternating external heat flow environment in orbit, causing a deflection in the camera attitude pointing. This deflection changes over time according to a certain pattern. Let the amplitude of the change in the Euler angle of the camera installation attitude be (δα1, δβ1, δγ1). In this implementation case, this change is considered to be composed of multiple frequency components with different amplitudes, which has the following representation:

[0145]

[0146] Where a i ,b i ,c i f represents the amplitude of each frequency component. i Indicates frequency, Let R represent the phase of each frequency component. Then the resulting camera attitude change matrix R... temp for,

[0147]

[0148] Then, considering structural deformation factors, the camera's orientation is,

[0149] C cam2 =C orb *R att *R cam2body *R temp

[0150] During satellite operation, some attitude control components on board will generate disturbances at a certain frequency, which will also affect the camera's attitude pointing. Let the equivalent attitude angle change of the camera caused by the micro-vibration of the components be (δα2, δβ2, δγ2). In this implementation case, it is represented in the following form:

[0151]

[0152] The camera attitude pointing change R caused by the disturbance vib for

[0153]

[0154] After considering the effects of disturbance, the camera attitude pointing is

[0155] C cam3 =C orb *R att *R cam2body *R temp *R vib

[0156] The third step involves calculating the imaging vector pointing of each pixel on the camera detector based on camera imaging parameters such as principal point and principal distance, and further considering the effect of the atmosphere on the imaging light.

[0157] The imaging ray of a camera consists of two parts: the starting position of the imaging ray and the imaging ray vector. This is based on the optical imaging geometry model (such as...). Figure 3 As shown), the imaging pixel center, optical imaging center, and ground object imaging position are located on the same straight line. Generally, the optical imaging center is taken as the starting position of the imaging ray. Based on the satellite orbital position calculated above and the camera imaging center's installation position on the satellite, the coordinate position of the camera imaging center at each moment [X] can be accurately calculated. s ,Y s Z s ].

[0158] Next, determine the direction of the imaging ray corresponding to each pixel. According to the geometric imaging model, let the camera focal length be f, the position of the camera principal point be (x0, y0) (in the camera coordinate system), and the position of the image point be (x, y). Then, the direction of the imaging ray corresponding to this image point is:

[0159]

[0160] Based on the camera orientation obtained above, the orientation of the imaging ray in the Earth coordinate system can be further calculated. The equation of the imaging ray can be expressed as:

[0161]

[0162] When the angle of tilted imaging is large, it is also necessary to consider the refraction of light by the atmosphere and calculate the final direction of the imaging light vector after atmospheric refraction according to the atmospheric refraction model.

[0163] The fourth step involves using a high-precision 3D scene model as the data base map to accurately calculate the intersection point between the imaging vector and the 3D scene model base map, thereby obtaining the camera's accurate imaging position.

[0164] Traditional planar image base maps cannot represent the geometric shapes of surface protrusions such as buildings. When simulating from an oblique perspective, they suffer from geometrical positioning errors and the inability to simulate building sides. This method uses a true 3D model that accurately represents surface undulations and feature shapes as the base map. The intersection of the imaging ray with the base map model is the true imaging geographic location, effectively avoiding the aforementioned problems or errors. The algorithm for calculating the intersection point of the imaging ray with the 3D model depends on the representation method used for the surface 3D model. Common 3D model representation methods include point cloud representation, triangular mesh representation, and voxel representation, as shown in Figures 4(a), 4(b), and 4(c). This implementation uses triangular mesh representation. The following algorithm can be used to calculate whether the imaging ray intersects with a triangular facet:

[0165] Let point P be a point on the triangular facet, and its coordinates be denoted as . The normal vector of the triangular facet is Then the parameters of the equation of the line corresponding to the intersection point If k0∈[0,+∞) and the intersection point is located inside the triangular facet, then the imaging ray intersects the triangular facet, and the coordinates of the intersection point are...

[0166]

[0167] To accurately calculate the actual imaging location, it is necessary to traverse and calculate the intersection points of all the constituent triangles in the triangular mesh model with the imaging ray, and also to determine the angle between the imaging ray and the normal of the imaging location area, as well as whether the intersection point is blocked by other ground features.

[0168] The fifth step involves comprehensively utilizing parameters such as the ground reflection characteristics, environmental radiation characteristics, and camera radiation imaging characteristics of each imaging geographic area to calculate the pixel value of each imaging point and simulate the formation of a simulation image.

[0169] The radiation energy of optical remote sensing imaging mainly comes from the radiation energy reflected by the ground imaging location area along the imaging direction. The energy level is mainly related to the area of ​​the ground imaging location area, the two-way reflectance distribution function (BRDF) of the ground objects at the imaging location, and the radiation intensity entering the imaging location area.

[0170] The image point corresponds to the imaging location area. Let the detector pixel size be d, the camera system focal length be f, and the satellite imaging altitude be h. Then, the projection of the imaging location area corresponding to each image point on the imaging ray normal plane is approximately a square area with a side length of hd / f. Further calculate the angle θ between the normal of the imaging location area and the imaging ray. The area of ​​the imaging location area is approximately hd / (fcos(θ)).

[0171] The bidirectional reflectance distribution function of the imaging location area is related to factors such as ground feature type, light source direction, and observation direction, and can be selected from the model database or determined based on experience.

[0172] The radiation energy reflected from the imaging location mainly includes solar radiation and radiation reflected from surrounding ground features to that location.

[0173] The solar radiation intensity of the imaging location area can be calculated using atmospheric radiative transfer models such as Modtran. For simulations at tilt angles, since many imaging areas are located on the sides of ground features, the corresponding solar radiation intensity must be calculated based on the direction of solar radiation, the normal of the ground features, and the observation direction. The shading effect of surrounding surface protrusions must also be considered, i.e., determining the visibility of the imaging location area to the sun. If the area is obscured by surrounding ground features and therefore invisible to the sun, this portion of solar radiation energy is not included. Specifically, visibility is determined by forming a sunlight vector from the imaging location point against the direction of solar illumination, and calculating whether this vector intersects with the 3D model of the ground features. If so, it is considered that the ground features at that location are blocking direct sunlight; otherwise, the imaging location can receive direct solar radiation energy. The intensity of directly reflected solar radiation is:

[0174]

[0175] In the formula The intensity of solar radiation reflected from the imaging location region. It is a two-way reflection distribution function. V represents the intensity of solar radiation received in the imaging location region. 0 The visibility of the sun at this location is 1 if visible and 0 if invisible. S is the range of the imaging location area.

[0176] The reflection intensity of the ground object's reflected radiation is closely related to the three-dimensional shape of the imaging area and the reflectivity of the ground object, and cannot be directly calculated analytically. However, it can be solved using ray tracing. The contribution of radiation energy reflected from different directions to the imaging location area to the imaging radiation intensity is as follows:

[0177]

[0178] in Let be the bidirectional reflection distribution coefficient of the reflected light from direction Ω. Let be the reflected radiation flux from direction Ω.

[0179] Therefore, the total radiation intensity reflected from this imaging location is:

[0180]

[0181] In the above formula This represents the radiation intensity reflected from a ground feature's reflection point, and its value can be recursively solved using the above calculation method. After traversing as many light directions as possible and the tracking depth as possible, the above formula gradually converges to the true radiation state. In the simulation process, in order to control the complexity and efficiency of the simulation calculation, and considering that the energy decays rapidly after multiple reflections and does not contribute much to the total energy entering the camera image, the Monte Carlo method is usually used to calculate the radiation intensity of the path tracked to a limited depth along a certain number of reflection directions.

[0182] In optical remote sensing imaging, due to the diffraction effect of the camera's optical system, the image of a point light source on the ground is represented by an Airy disk distributed over a region on the image plane. Its energy distribution function is called the point spread function, and its shape is as follows: Figure 5 As shown. The diffraction effect of an optical system causes the energy reflected from the imaging location region of a single image point to radiate to surrounding pixels. Similarly, the image point also receives radiated energy reflected from the imaging location regions of surrounding pixels. Therefore, the radiated energy ultimately received by a certain image point is also affected by the radiated energy reflected from the surrounding region of its imaging location, and has the following form:

[0183]

[0184] In the formula, f(x,y) is the radiance at image point (x,y), L(xu,yv) is the radiance of image points within the region that contributes to the radiant energy of that image point, and h(u,v) is the normalized point spread function.

[0185] Based on the radiation intensity of a single image point, and taking into account characteristics such as detector noise and quantum efficiency, the DN value of that image point can be calculated.

[0186] For high-resolution optical remote sensing imaging, due to the high speed of satellite flight, in order to ensure the radiation energy required for the photoelectric conversion of the detector, a linear array pushbroom imaging method is usually adopted to acquire multiple imaging results of the same ground object at different times, and then the images are superimposed and processed accordingly to form the final remote sensing image.

[0187] This embodiment also provides a full-view optical remote sensing image simulation system, which includes: a first module for obtaining the imaging position of the satellite at a specified time based on the satellite orbital dynamics model; a second module for obtaining the orbital coordinate system pointing of the satellite orbital position at a specified time based on the satellite orbital dynamics model, obtaining the satellite attitude matrix based on the satellite attitude information, obtaining the satellite attitude pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite orbital position at a specified time and the satellite attitude matrix, and obtaining the camera pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite orbital position at a specified time, the satellite attitude matrix, the imaging camera installation matrix, the installation attitude change matrix, and the camera attitude pointing change caused by micro-vibration; a third module for obtaining the imaging vector pointing of each pixel on the camera detector based on the camera pointing in the inertial coordinate system; a fourth module for obtaining the camera imaging geographical location based on the intersection of the imaging vector pointing of each pixel on the camera detector and a preset three-dimensional scene model base map; and a fifth module for obtaining the radiation intensity of each imaging point based on the camera imaging geographical location, and obtaining the pixel value of each imaging point based on the radiation intensity, detector noise, and quantum efficiency of each imaging point.

[0188] This embodiment models the entire process of acquiring remote sensing images, including orbit positioning, attitude determination, camera imaging, and atmospheric transmission, and their error characteristics, based on the actual physical process of remote sensing image acquisition. It realistically considers the impact of each stage of the imaging platform and remote sensor on the quality of the simulated image. This embodiment uses a real-world 3D model of the scene as the simulation data source, accurately reflecting the terrain undulations and 3D shapes of surface features. It can particularly accurately describe the facade information of surface protrusions such as buildings, realistically simulating the mutual occlusion of buildings and other surface protrusions in tilted-angle imaging. This embodiment employs a ray-tracing radiation simulation method, comprehensively considering the influence of major environmental light sources such as solar radiation and ground feature reflection on the simulated image, effectively solving the problem of image simulation in shadow areas and achieving high-fidelity imaging simulation of ground features.

[0189] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. A full-view optical remote sensing image simulation method, characterized in that... include: The imaging position of the satellite at a specified time is obtained based on the satellite orbital dynamics model; The orbital coordinate system pointing of the satellite at a specified time is obtained from the satellite orbital dynamics model. The satellite attitude matrix is ​​obtained from the satellite attitude information. The orientation of the satellite attitude in the inertial coordinate system is obtained from the orbital coordinate system pointing of the satellite at a specified time and the satellite attitude matrix. The orientation of the camera in the inertial coordinate system is obtained from the orbital coordinate system pointing of the satellite at a specified time, the satellite attitude matrix, the imaging camera installation matrix, the installation attitude change matrix, and the camera attitude orientation change caused by micro-vibration. The imaging vector pointing of each pixel on the camera detector is obtained based on the camera's orientation in the inertial coordinate system. The geographical location of the camera image is obtained by the intersection of the imaging vector of each pixel on the camera detector and the preset 3D scene model base map. The radiation intensity of each imaging point is obtained based on the geographical location of the camera imaging, and the pixel value of each imaging point is obtained based on the radiation intensity, detector noise, and quantum efficiency of each imaging point.

2. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: The satellite orbital dynamics model is obtained through the following formula: in, The acceleration of the satellite in the J2000; The acceleration due to gravity at the Earth's center; This is the acceleration due to the non-spherical gravitational effect of the Earth.

3. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: Satellite attitude matrix R att It can be obtained through the following formula: Among them, R att For the satellite attitude matrix, Let ω be the Euler angle of the satellite attitude about the X-axis, ω be the Euler angle of the satellite attitude about the Y-axis, and κ be the Euler angle of the satellite attitude about the Z-axis.

4. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: The orientation of the satellite in the inertial coordinate system, C att It can be obtained through the following formula: C att =C orb *R att ; Among them, C att C represents the orientation of the satellite in the inertial coordinate system. orb R is the orbital coordinate system pointing to the satellite's orbital position at a specified time. att For the satellite attitude matrix; Install attitude change matrix R temp It can be obtained through the following formula: Wherein, δα1 is the change of Euler angles about the X-axis due to the change of camera mounting attitude on track, δβ1 is the change of Euler angles about the Y-axis due to the change of camera mounting attitude on track, and δγ1 is the change of Euler angles about the Z-axis due to the change of camera mounting attitude on track.

5. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: Camera attitude pointing change R caused by micro-vibration vib It can be obtained through the following formula: Wherein, δα2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the X-axis, δβ2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the Y-axis, and δγ2 is the change of the Euler angle of the camera attitude caused by the micro-vibration about the Z-axis.

6. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: The camera's orientation C in the inertial coordinate system cam3 It can be obtained through the following formula: C cam3 =C orb *R att *R cam2body *R temp *R vib ; Among them, C orb R is the orbital coordinate system pointing to the satellite's orbital position at a specified time. att R is the satellite attitude matrix. temp To install the attitude change matrix, R vib R represents the camera's attitude pointing change caused by micro-vibrations. cam2body Install a matrix for the imaging camera.

7. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: The imaging vector pointing to each pixel on the camera detector is obtained by the following formula: Among them, C cam3 Here, X represents the camera's orientation in the inertial coordinate system, Y represents the pixel's imaging position in the inertial coordinate system, and Z represents the pixel's imaging position in the inertial coordinate system. s Let X and Y be the imaging position of the camera center pixel in the inertial coordinate system. s Let Y and Z be the Y-coordinates of the imaging position of the camera's center pixel in the inertial coordinate system. s Let X be the Z-coordinate of the imaging position of the camera center pixel in the inertial coordinate system, k be a constant coefficient, x′ be the X-axis component of the imaging vector of the pixel in the camera coordinate system, y′ be the Y-axis component of the imaging vector of the pixel in the camera coordinate system, and z′ be the Z-axis component of the imaging vector of the pixel in the camera coordinate system. d Let X be the component of the imaging vector of a pixel in the inertial coordinate system in the X direction, and Y be the component of the imaging vector in the inertial coordinate system in the X direction. d Let Z be the component of the imaging vector of a pixel in the inertial coordinate system in the Y direction. d The component of the imaging vector of a pixel in the inertial coordinate system in the Z direction.

8. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: The geographic location of the camera image is obtained using the following formula: Where X1 is the X-axis coordinate of the camera's image geographic location, Y1 is the Y-axis coordinate of the camera's image geographic location, Z1 is the Z-axis coordinate of the camera's image geographic location, and X... s Let X and Y be the imaging position of the camera center pixel in the inertial coordinate system. s Let Y and Z be the Y-coordinates of the imaging position of the camera's center pixel in the inertial coordinate system. s Let X be the Z-coordinate of the camera center pixel's imaging position in the inertial coordinate system, k0 be the second coefficient, and X be the Z-coordinate of the imaging position in the inertial coordinate system. d Let X be the component of the imaging vector of a pixel in the inertial coordinate system in the X direction, and Y be the component of the imaging vector in the inertial coordinate system in the X direction. d Let Z be the component of the imaging vector of a pixel in the inertial coordinate system in the Y direction. d The component of the imaging vector of a pixel in the inertial coordinate system in the Z direction.

9. The full-view optical remote sensing image simulation method according to claim 1, characterized in that: The radiation intensity at each imaging point is obtained using the following formula: f(x,y)=∑ u ∑ v L(x-u,y-v)h(u,v); Where f(x,y) is the radiation intensity at the imaging point (x,y), x is the x-axis coordinate of the imaging point, y is the y-axis coordinate of the imaging point, L(xu,yv) is the radiation intensity of the image point within the region that contributes to the radiation energy of the imaging point, h(u,v) is the normalized point spread function, u is the x-axis range of the region that contributes to the radiation energy of the imaging point, and v is the y-axis range of the region that contributes to the radiation energy of the imaging point.

10. A full-view optical remote sensing image simulation system, characterized in that... include: The first module is used to obtain the satellite's imaging position at a specified time based on the satellite orbital dynamics model; The second module is used to obtain the orbital coordinate system pointing of the satellite orbital position at a specified time based on the satellite orbital dynamics model, obtain the satellite attitude matrix based on the satellite attitude information, obtain the satellite attitude pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite orbital position at a specified time and the satellite attitude matrix, and obtain the camera pointing in the inertial coordinate system based on the orbital coordinate system pointing of the satellite orbital position at a specified time, the satellite attitude matrix, the imaging camera mounting matrix, the mounting attitude change matrix, and the camera attitude pointing change caused by micro-vibration. The third module is used to obtain the imaging vector pointing of each pixel on the camera detector based on the camera's pointing in the inertial coordinate system. The fourth module is used to obtain the geographical location of the camera image based on the intersection of the imaging vector pointing to each pixel on the camera detector and the preset 3D scene model base map. The fifth module is used to obtain the radiation intensity of each imaging point based on the geographical location of the camera imaging, and to obtain the pixel value of each imaging point based on the radiation intensity, detector noise and quantum efficiency of each imaging point.

Images