A method for determining coordinates of a lunar observation remote sensing image

Abstract

The application discloses a kind of coordinate determination methods of lunar observation remote sensing image, steps are as follows: the three-dimensional coordinates of each pixel point in lunar image are established;Determine the satellite view angle libration based on the lunar observation position of satellite;The azimuth between any two points on the moon is determined according to the geographical coordinates of the two points, the projection coordinates of two latitude and longitude coordinate points on the image plane and the azimuth thereof are determined;The angle relationship between the scanning direction when satellite sampling and the lunar fixed coordinate system is determined;The geographical coordinates of the observed lunar image under the lunar fixed coordinate system are determined.The method of the application can avoid the complex imaging model based on rigorous geometric relationship, and can solve the deficiency that the image matching method has strong dependence on image gray or texture information;Only a small amount of image processing and simple coordinate conversion can establish the accurate mapping relationship between image coordinate and moon surface coordinate.

Description

Technical Field

[0001] This invention belongs to the field of lunar remote sensing imaging and observation technology, and relates to a method for geometrically locating lunar images obtained by lunar spectral measurements performed by instruments in near-Earth space, specifically a method for determining the coordinates of lunar observation remote sensing images. Background Technology

[0002] The Moon, the closest natural body to Earth, is widely considered a stable solar diffuser. Ground-based instruments and remote sensing instruments operating in all Earth orbits can regularly observe the Moon, and its radiation energy falls within the dynamic range of most payloads. Using the Moon as a calibration reference for on-orbit instruments offers unique advantages such as low observation costs, high stability, and absence of atmospheric interference. It also provides ample opportunities for cross-calibration, even between instruments with non-overlapping observation times. Using the Moon as a high-precision calibration reference requires considering its directional reflectivity, which is closely related to the geometric relationship between the Sun, Moon, and satellite. Current lunar-based calibration techniques focus on the global radiation characteristics of the near side of the Moon, with the commonly used ROLO model providing an equivalent disk simulation value as a reference for satellite-measured lunar irradiance. This calibration method has an absolute uncertainty of approximately 5-10%, and its relative accuracy is affected by lunar phase angle and libration modeling residuals. Pieters et al. proposed using the radiance / reflectivity of specific regions on the lunar surface as an optical standard for instruments and recommended several calibration areas, all covering an area of ​​approximately 200 km. Geostationary satellites and high-resolution satellites operating in polar orbits are expected to calibrate instruments based on the radiation characteristics of small areas on the Moon. Such practices and applications are based on the geographical location of lunar observation images, and clearly defining the pixel location of remote sensing images is usually a basic prerequisite for data application.

[0003] Because lunar scanning by Earth-orbiting satellites often involves maneuvers, constructing a rigorous imaging geometry model involves complex coordinate transformations. Furthermore, these satellites are designed for Earth observation and have corresponding scanning mechanisms, but imaging the Moon from a distance requires consideration of more complex factors (such as stellar phase differences), and the azimuth parameters of high-resolution satellites are often unavailable. In addition, using general sensor models such as rational polynomial models is difficult and has limited accuracy. When using image matching methods, the Moon always lacks sufficient grayscale or texture information at high phase angles. Due to libration, the visible lunar surface is always slightly different, but it has a special rotational relationship with the nominal lunar orthophoto image centered on the lunar latitude and longitude origin (latitude and longitude between -90° and 90°). Therefore, based on these characteristics, it is necessary to develop a method for calculating lunar surface coordinates in lunar images that can serve the radiometric correction of satellite instruments. Summary of the Invention

[0004] To address the shortcomings of the existing technologies, the present invention aims to provide a method for determining the coordinates of lunar remote sensing images, establishing a relationship between lunar image coordinates and the actual lunar position. This method avoids complex imaging models based on strict geometric relationships and overcomes the shortcomings of image matching methods that rely heavily on image grayscale or texture information. Only a small amount of image processing and simple coordinate transformation are required to establish an accurate mapping relationship between image coordinates and lunar surface coordinates.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] The present invention provides a method for determining the coordinates of a lunar remote sensing image, comprising the following steps:

[0007] 1) Sample lunar images acquired by different satellite instruments into a circular moon, and determine the lunar parameters through image processing; based on the resampled images, establish the three-dimensional coordinates of each pixel point in the lunar image based on the pixel position in the image plane coordinate system;

[0008] 2) Determine the libration from the satellite's perspective based on the satellite's lunar observation position;

[0009] 3) Calculate the azimuth angle based on the geographic coordinates of any two points on the lunar surface to determine the azimuth between the two points. Based on the spatial collinearity between the image point and the object point, determine the projected coordinates of the two latitude and longitude coordinate points on the image plane and their azimuth angles.

[0010] 4) Determine the angular relationship between the scanning direction of the satellite during sampling and the lunar-fixed coordinate system based on the two azimuth angles obtained in step 3);

[0011] 5) Based on the results determined in steps 2) and 4), and the three-dimensional coordinates established in step 1), determine the geographic coordinates of the observed lunar image in the lunar-fixed coordinate system.

[0012] Further, step 1) specifically includes:

[0013] The lunar observation was conducted using a linear array multispectral imager through on-orbit attitude maneuvers, scanning one frame at a time. After one complete maneuver, a two-dimensional image containing the Moon was formed. Based on the scanning principle, the observed image was sampled to obtain an equivalent lunar disk image. A bilateral filtering algorithm was used to smooth lunar surface noise while preserving sharp lunar edges, and lunar contour points were extracted. A least-squares fitting method was used to fit the equivalent lunar disk, determining the center position of the lunar disk in pixels (x) in the image plane coordinate system. center ,y center ) and radius r; based on the image row and column index (i) of the image plane coordinate system image ,j image Calculate the nominal rectangular coordinates of each pixel in the region containing the moon;

[0014] y = -(j image -y center )

[0015] z = i image -z center

[0016]

[0017] in, To solve for the three-dimensional coordinates of each pixel, we obtain the nominal rectangular coordinates of each pixel under the nominal lunar orthophoto image.

[0018] Furthermore, step 2) specifically includes:

[0019] When the Moon is not in libration, the vector from the Moon's center to its apparent center coincides with the +X axis of the lunar-fixed coordinate system, pointing towards the average Earth direction, while the +Z axis points towards the Moon's average rotation direction. At this time, the satellite's geographic coordinates on the lunar surface are (0, 0). Using the satellite's observation viewpoint as a reference, and taking the sampling time t during the scanning of the Moon's center, the Moon's libration relative to the satellite is calculated, determining the rectangular coordinates of the intersection point of the line connecting the satellite and the Moon's center with the lunar surface. Then, based on the relationship between the lunar geographic coordinate system and the lunar-fixed coordinate system, the satellite's geographic coordinates on the lunar surface are calculated (lon). sub-satellite ,lat sub-satellite ), which is the coordinate of the satellite's sub-satellite point relative to the moon, and also the libration in the longitude and latitude directions.

[0020] Furthermore, step 3) specifically includes:

[0021] 31) Take any two points on the near side of the Moon. Let the first point be the projection of the satellite onto the lunar surface, located at the apparent center of the Moon, denoted as point A, and its corresponding geographic coordinates be (lon). sub-satellite ,lat sub-satellite ) or (lon A ,lat A ); Take the coordinates of the second point with the same latitude as the first point and a longitude increment of 2° as (lon B ,lat B Let B be a point, where lon B =lon A +2, lat B =lat A ; Calculate the azimuth angle of the line connecting points A and B, with the north direction of the lunar coordinate system as the reference, starting from the lunar projection point A of the satellite.

[0022] Δlon=lon B -lon A

[0023] X = cos(lat) B sin(Δlon)

[0024] Y = cos(lat) A sin(lat) B )-sin(lat A cos(lat) B cos(Δlon)

[0025]

[0026] Where Δlon is the difference in longitude between the two coordinate points;

[0027] 32) Determine the transformation between the instrument-fixed coordinate system and the satellite body coordinate system based on the installation relationship between the multispectral imager and the satellite body coordinate system.

[0028] 33) Based on the quaternion attitude (q0, q1, q2, q3) corresponding to the central scan line when the satellite scans the moon. T Determine the transformation relationship between the satellite's body coordinate system and the Earth's inertial coordinate system:

[0029]

[0030] in, q0 is the transformation matrix from the satellite body coordinate system to the Earth inertial coordinate system; q0 and (q1,q2,q3) are the real and imaginary parts of the attitude quaternion, respectively.

[0031] 34) Combine the lunar geographic coordinates (lon) of the two points in step 31). A ,lat A ) and (lon B ,lat B Transform them respectively to Cartesian coordinates in the lunar fixed coordinate system. and Based on ephemeris data, and through coordinate transformation, the unit vectors of the satellite at the two lunar surface coordinate points at the observation time in the lunar inertial coordinate system were obtained. and (The coordinate axes of the lunar inertial coordinate system and the Earth inertial coordinate system are similar and do not require conversion); then, based on the instrument fixed coordinate system and satellite body coordinate system obtained in steps 32) and 33), and the conversion relationship between the satellite body coordinate system and the Earth inertial coordinate system, the line-of-sight vectors of the above two points in the instrument fixed coordinate system are obtained:

[0032]

[0033] in, These are rectangular coordinates in the instrument's fixed coordinate system. These are rectangular coordinates in the Earth's inertial coordinate system.

[0034] 35) Assuming the imaging of the linear multispectral imager is a single-center projection of a planar CCD, calculate the positions of points A and B in the image plane coordinate system from step 31) based on the spatial collinearity between the central projected image point, the object point, and the projection center; determine the pixel coordinates in the image plane coordinate system based on the line-of-sight vector in the instrument-fixed coordinate system obtained in step 34). That is: the pixel coordinates of the satellite's projection point A on the lunar surface and another coordinate point B at the same latitude; due to potential errors in steps 32) and 33), the lunar center coordinates (x) obtained by fitting in step 1) are used. center ,y center Correct the position of the satellite's projection point on the lunar surface and another coordinate point at the same latitude on the image plane;

[0035]

[0036] in, Let K be the coordinates of the center pixel of the image plane, K be the transformation matrix between the physical imaging coordinates and the image coordinate system, and f be the equivalent focal length of the instrument.

[0037] 36) Based on the results of step 35), calculate the azimuth angle of the line connecting the projection points of two points A and B on the lunar surface onto the image plane, with the north direction of the image as the reference.

[0038] Furthermore, step 4) specifically includes:

[0039] According to the definitions of points A and B, the azimuth angle... This represents the angular relationship between two points on the lunar surface and the north direction in the lunar-fixed coordinate system. This represents the relationship between the line connecting two points on the lunar surface to their projection points on the image plane and the vertical axis, based on the difference between the two angles. The angle between the scanning direction during satellite sampling and the lunar-fixed coordinate system is determined to represent the rotation of the actual lunar observation image relative to the nominal image.

[0040] Furthermore, step 5) specifically includes:

[0041] 51) Based on the three-dimensional rectangular coordinates (x, y, z) of the pixels on the lunar disk in step 1), T The latitude, longitude, libration, and rotation angle information of the satellite relative to the moon obtained in steps 2) and 4) are used to determine the geographic coordinates of the pixels in the observed image.

[0042]

[0043] in, The nominal rectangular coordinates of the lunar pixels obtained in step 1); The actual rectangular coordinates of the lunar pixels in the observed image; φ, Θ, and ψ represent the longitude, latitude, libration, and rotation relative to the nominal lunar image of the observed image, respectively. sub-satellite lat sub-satellite and The transformation matrix is ​​defined as follows:

[0044]

[0045]

[0046]

[0047] 52) Based on the rectangular coordinates of the observed image pixels Determine the lunar geographic coordinates (lon, lat) of the corresponding pixel, and obtain the lunar latitude and longitude of the actual observed image:

[0048]

[0049] or

[0050] The beneficial effects of this invention are:

[0051] This invention does not rely on texture or grayscale features in image matching methods. It uses minimal image processing and simple coordinate transformation to map observed lunar image pixels to a specified lunar coordinate system, making it suitable for any lunar phase observation where the moon is illuminated.

[0052] This invention forms the basis for selecting the location of lunar surface radiation calibration fields and studying their photometric characteristics, and can be widely applied to lunar observation images in near-Earth space. Attached Figure Description

[0053] Figure 1 This is a flowchart of the method of the present invention.

[0054] Figure 2 This is a schematic diagram of the nominal coordinate system based on the lunar disk.

[0055] Figure 3 This is a schematic diagram of the spatial relationship between object points and image points.

[0056] Figure 4a This is a schematic diagram of a lunar image (sampled as an equivalent lunar disk) observed by a multispectral imager.

[0057] Figure 4b for Figure 4aA schematic diagram of an image after geocoding based on the determined coordinates. Detailed Implementation

[0058] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0059] Reference Figure 1 As shown, the method for determining the coordinates of a lunar remote sensing image according to the present invention comprises the following steps:

[0060] 1) Resample lunar images acquired by different satellite instruments into a circular lunar shape, and determine the lunar parameters through image processing; based on the resampled images and the pixel positions in the image plane coordinate system, establish the three-dimensional coordinates of each pixel point in the lunar image; specifically including:

[0061] The lunar observation was conducted using a linear array multispectral imager through on-orbit attitude maneuvers, scanning one frame at a time. After one complete maneuver, a two-dimensional image containing the Moon was formed. Based on the scanning principle, the observed image was sampled to obtain an equivalent lunar disk image. A bilateral filtering algorithm was used to smooth lunar surface noise while preserving sharp lunar edges, and lunar contour points were extracted. A least-squares fitting method was used to fit the equivalent lunar disk, determining the center position of the lunar disk in pixels (x) in the image plane coordinate system. center ,y center ) and radius r; based on the image row and column index (i) of the image plane coordinate system image ,j image Calculate the nominal rectangular coordinates of each pixel in the region containing the moon;

[0062] y = -(j image -y center )

[0063] z = i image -z center

[0064]

[0065] in, To solve for the three-dimensional coordinates of each pixel, we obtain the nominal rectangular coordinates of each pixel under the nominal lunar orthophoto image.

[0066] 2) Determining the libration from the satellite's perspective based on the satellite's lunar observation position; specifically including:

[0067] When the Moon is not in libration, the vector from the Moon's center to its apparent center coincides with the +X axis of the lunar-fixed coordinate system, pointing towards the average Earth direction, while the +Z axis points towards the Moon's average rotation direction. In this case, the satellite's geographic coordinates on the lunar surface projection point are (0, 0). In most cases, the libration is not zero. Using the satellite's observation viewpoint as a reference, and taking the sampling time t when scanning the Moon's center, the Moon's libration relative to the satellite is calculated. The rectangular coordinates of the intersection point of the line connecting the satellite and the Moon's center with the lunar surface are then determined. Finally, based on the relationship between the lunar geographic coordinate system and the lunar-fixed coordinate system, the satellite's geographic coordinates on the lunar surface projection point are calculated. sub-satellite ,lat sub-satellite ), which is the coordinate of the satellite's sub-satellite point relative to the moon, and also the libration in the longitude and latitude directions.

[0068] 3) Calculate the azimuth angle based on the geographic coordinates of any two points on the lunar surface to determine the azimuth between the two points, referring to... Figure 3 As shown, the spatial collinearity between the image point and the object point determines the projected coordinates and azimuth angles of the two latitude and longitude coordinate points on the image plane; specifically including:

[0069] 31) Take any two points on the near side of the Moon. The first point is the projection of the satellite onto the lunar surface, located at the apparent center of the Moon, denoted as point A, and its corresponding geographic coordinates are (lon). sub-satellite ,lat sub-satellite ) or (lon A ,lat A ); Take the coordinates of the second point with the same latitude as the first point and a longitude increment of 2° as (lon B ,lat B Let B be a point, where lon B =lon A +2, lat B =lat A ; Calculate the azimuth angle of the line connecting points A and B, with the north direction of the lunar coordinate system as the reference, starting from the lunar projection point A of the satellite.

[0070] Δlon=lon B -lon A

[0071] X = cos(lat) B sin(Δlon)

[0072] Y = cos(lat) A sin(lat) B )-sin(lat A cos(lat) B cos(Δlon)

[0073]

[0074] Where Δlon is the difference in longitude between the two coordinate points;

[0075] 32) Determine the transformation between the instrument-fixed coordinate system and the satellite body coordinate system based on the installation relationship between the multispectral imager and the satellite body coordinate system.

[0076] 33) Based on the quaternion attitude (q0, q1, q2, q3) corresponding to the central scan line when the satellite scans the moon. T Determine the transformation relationship between the satellite's body coordinate system and the Earth's inertial coordinate system:

[0077]

[0078] in, q0 is the transformation matrix from the satellite body coordinate system to the Earth inertial coordinate system; q0 and (q1,q2,q3) are the real and imaginary parts of the attitude quaternion, respectively.

[0079] 34) Combine the lunar geographic coordinates (lon) of the two points in step 31). A ,lat A ) and (lon B ,lat B Transform them respectively to Cartesian coordinates in the lunar fixed coordinate system. and Based on ephemeris data, and through coordinate transformation, the unit vectors of the satellite at the two lunar surface coordinate points at the observation time in the lunar inertial coordinate system were obtained. and The coordinate axes of the lunar inertial coordinate system and the Earth inertial coordinate system are similar and do not require transformation; then, based on the instrument-fixed coordinate system and satellite body coordinate system obtained in steps 32) and 33), and the transformation relationship between the satellite body coordinate system and the Earth inertial coordinate system, the line-of-sight vectors of the two points mentioned above in the instrument-fixed coordinate system are obtained:

[0080]

[0081] in, These are rectangular coordinates in the instrument's fixed coordinate system. These are rectangular coordinates in the Earth's inertial coordinate system.

[0082] 35) Assuming the imaging of the linear multispectral imager is a single-center projection of a planar CCD, calculate the positions of points A and B in the image plane coordinate system from step 31) based on the spatial collinearity between the central projected image point, the object point, and the projection center; determine the pixel coordinates in the image plane coordinate system based on the line-of-sight vector in the instrument-fixed coordinate system obtained in step 34). That is: the pixel coordinates of the satellite's projection point A on the lunar surface and another coordinate point B at the same latitude; due to potential errors in steps 32) and 33), the lunar center coordinates (x) obtained by fitting in step 1) are used. center ,y center Correct the position of the satellite's projection point on the lunar surface and another coordinate point at the same latitude on the image plane;

[0083]

[0084] in, Let K be the coordinates of the center pixel of the image plane, K be the transformation matrix between the physical imaging coordinates and the image coordinate system, and f be the equivalent focal length of the instrument.

[0085] 36) Based on the results of step 35), calculate the line connecting the projection points of two points A and B on the lunar surface onto the image plane in the north direction of the image ( Figure 2 The azimuth angle with the +Z axis direction as the reference.

[0086] 4) Based on the two azimuth angles obtained in step 3), determine the angular relationship between the satellite sampling direction and the lunar-fixed coordinate system (here, the Earth-meaning / rotational axis coordinate system is used); specifically including:

[0087] According to the definitions of points A and B, the azimuth angle... This represents the angular relationship between two points on the lunar surface and the north direction in the lunar-fixed coordinate system. This represents the relationship between the line connecting two points on the lunar surface to their projection points on the image plane and the vertical axis, based on the difference between the two angles. The angle between the scanning direction during satellite sampling and the lunar-fixed coordinate system is determined to represent the rotation of the actual lunar observation image relative to the nominal image.

[0088] 5) Based on the results determined in steps 2) and 4), and the three-dimensional coordinates established in step 1), determine the geographic coordinates of the observed lunar image in the lunar-fixed coordinate system; specifically including:

[0089] 51) Based on the three-dimensional rectangular coordinates (x, y, z) of the pixels on the lunar disk in step 1), T The latitude, longitude, libration, and rotation angle information of the satellite relative to the moon obtained in steps 2) and 4) are used to determine the geographic coordinates of the pixels in the observed image.

[0090]

[0091] in, The nominal rectangular coordinates of the lunar pixels obtained in step 1); The actual rectangular coordinates of the lunar pixels in the observed image; φ, Θ, and ψ represent the longitude, latitude, libration, and rotation relative to the nominal lunar image of the observed image, respectively. sub-satellite lat sub-satellite and The transformation matrix is ​​defined as follows:

[0092]

[0093]

[0094]

[0095] 52) Based on the rectangular coordinates of the observed image pixels Determine the lunar geographic coordinates (lon, lat) of the corresponding pixel, and obtain the lunar latitude and longitude of the actual observed image:

[0096]

[0097] or

[0098] Figure 4a and Figure 4b The lunar images observed by the multispectral camera 2 of Jilin-1 Spectrum 02 satellite during its on-orbit maneuver, along with the geographic coordinates of lunar pixel points calculated based on the method and process of this invention, are represented by geocoded lunar images.

[0099] This invention has many specific applications. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A method for determining coordinates of a remote sensing image of a lunar observation, characterized in that, The steps are as follows: 1) sampling the moon images collected by different satellite instruments into a circular moon, determining the image plane moon parameters through an image processing process; based on the resampled images, establishing the three-dimensional coordinates of each pixel point in the moon images based on the pixel position in the image plane coordinate system; 2) determining the libration of the satellite viewing angle based on the satellite lunar observation position; 3) Calculate the azimuth angle from the geographic coordinates of any two points on the lunar surface to determine the orientation between the two points, based on the spatial collinearity of the image point and object point, determine the projection coordinates of the two latitude and longitude coordinate points on the image plane and their azimuth angle Any two points on the lunar surface, the first point is the satellite projection point on the lunar surface, located in the apparent center of the moon, recorded as point A, corresponding geographical coordinates (lon sub-satellite ,lat sub-satellite ) or (lon A ,lat A ); the second point is taken with the same latitude and longitude increment of 2° of the first point, recorded as point B; the azimuth of the line connecting points A and B is calculated with the lunar fixed coordinate system north direction as reference 4) determining the angle relationship between the scanning direction of the satellite sampling and the moon fixed coordinate system based on the two azimuth angles obtained in step 3); 5) determining the geographic coordinates of the observed moon images in the moon fixed coordinate system based on the determination results of steps 2) and 4), and the three-dimensional coordinates established in step 1); The step 4) specifically comprises: According to the definition of points A and B, the azimuth angle represents the angle relationship between two points on the lunar surface and the north direction of the lunar fixed coordinate system, represents the relationship between the line connecting the two points on the lunar surface and the longitudinal axis of the projection point in the image plane, and is determined according to the difference between the two angles determines the angle between the scanning direction at the time of satellite sampling and the lunar fixed coordinate system, and is used to represent the rotation of the actual observation image of the moon relative to the nominal image; The step 5) specifically comprises: 51) Determine the geographic coordinate position of the image element in the observation image according to the three-dimensional rectangular coordinates (x, y, z) of the image element on the moon disk in step 1), and the longitude and latitude libration and rotation angle information of the satellite relative to the moon obtained in steps 2) and 4). T 51) Determine the geographic coordinate position of the image element in the observation image according to the three-dimensional rectangular coordinates (x, y, z) of the image element on the moon disk in step 1), and the longitude and latitude libration and rotation angle information of the satellite relative to the moon obtained in steps 2) and 4). wherein, is the nominal rectangular coordinates of the lunar surface pixel obtained in step 1); is the actual rectangular coordinates of the observed image lunar surface pixel; φ, Θ and ψ represent the longitude, latitude libration and rotation of the observed image with respect to the nominal lunar image, respectively, lon sub-satellite , lat sub-satellite and The conversion matrix is defined as follows: 52) Rectangular coordinates of the observed image pixel Determination of the lunar geographic coordinates (lon, lat) of the corresponding pixel, acquisition of the lunar longitude and latitude of the actual observed image:

2. The method according to claim 1, wherein, The step 1) specifically comprises: The line array multispectral imager is used for observing the moon by on-orbit attitude maneuvering, one frame is scanned each time, and one two-dimensional image containing the moon is formed after one complete maneuvering; the observation image is sampled to obtain an equivalent moon disc image according to the scanning principle; the bilateral filtering algorithm is used to smooth the moon surface noise and retain the sharp edge of the moon, the moon outline points are extracted, the least square fitting method is used to fit the equivalent moon disc, and the pixel unit moon disc center position (x center ,y center ) and radius r in the image plane coordinate system are determined; the nominal rectangular coordinates of each pixel in the moon area are calculated according to the image row and column indexes (i image ,j image ) in the image plane coordinate system; y = -(j image - y center ) z = i image - z center wherein, for each pixel point of the solution, the three-dimensional coordinates of each pixel in the corresponding nominal lunar orthographic image.

3. The method according to claim 2, wherein, The step 2) specifically comprises: When the libration of the moon does not exist, the vector from the moon center to the apparent center of the moon coincides with the +X axis direction of the moon fixed coordinate system, points to the average earth direction, and the +Z axis points to the average rotation direction of the moon. At this time, the geographic coordinates of the satellite at the moon surface projection point are (0, 0); taking the sampling time t when the moon center is scanned as a reference, the libration of the moon relative to the satellite is calculated, and the right-angle coordinates of the intersection point between the line connecting the satellite and the moon center and the moon surface are determined; then, according to the relationship between the moon geographic coordinate system and the moon fixed coordinate system, the geographic coordinates (lon sub-satellite , lat sub-satellite ) of the satellite at the moon surface projection point, i.e. the sub-satellite point coordinates of the satellite relative to the moon, are calculated, which are also the longitude and latitude directions of the libration.

4. The method according to claim 3, wherein, The step 3) specifically comprises: 31) Take two points on the front of the moon, set the first point as the satellite projection point on the moon, located in the apparent center of the moon, recorded as point A, corresponding geographical coordinates (lon sub-satellite ,lat sub-satellite ) or (lon A ,lat A ); Take the second point with the same latitude as the first point, the longitude increment is 2°, the coordinates of the second point are (lon B ,lat B ), recorded as point B, wherein, lon B =lon A +2, lat B =lat A ; Take the satellite projection point A on the moon as the starting point to calculate the azimuth angle of the line connecting points A and B with the north direction of the moon fixed coordinate system as the reference Δlon = lon B -lon A X = cos(lat B ) sin(Δlon) Y = cos(lat A ) sin(lat B ) - sin(lat A ) cos(lat B ) cos(Alon) Wherein, Δlon is the longitude difference of the two coordinate points. 32) determining a conversion between the instrument-fixed coordinate system based on the multispectral imager and the satellite body coordinate system based on the mounting relationship between the multispectral imager and the satellite body coordinate system 33) The quaternion attitude (q0, q1, q2, q3) corresponding to the central scan line when the satellite scans the Moon T determining the conversion relationship between the satellite body coordinate system and the Earth inertial coordinate system: wherein, is the transformation matrix from the satellite body coordinate system to the earth inertial coordinate system; q0and (q1, q2, q3) are the real and imaginary parts of the attitude quaternion, respectively; 34) Convert the two points' selenographic coordinates (lon A ,lat A ) and (lon B ,lat B ) in step 31) to Cartesian coordinates (x ,y ) in the Moon-fixed coordinate system respectively. Based on the ephemeris data, through coordinate transformation, get the unit vectors of the satellite to the above two points on the Moon at the observation time in the lunar inertial coordinate system respectively According to the instrument-fixed coordinate system and the satellite body coordinate system obtained in steps 32) and 33), and the conversion relationship between the satellite body coordinate system and the Earth inertial coordinate system, get the line-of-sight vectors of the above two points in the instrument-fixed coordinate system: wherein, is a direct rectangular coordinate in the instrument-fixed coordinate system; is a direct rectangular coordinate in the earth inertial coordinate system; 35) The line array multi-spectral imager imaging assumption is single center projection of area array CCD, according to the space collinear relation of center projection image point, object point and projection center, the position of A, B two points in image plane coordinate system in step 31) is calculated; according to the line of sight vector in instrument fixed coordinate system obtained in step 34), the image element coordinate in image plane coordinate system is determined That is: the image element coordinate of satellite projection point A on the moon and another coordinate point B with same latitude; due to the potential error of step 32) and step 33), the position of satellite projection point on the moon and another coordinate point with same latitude on the image plane is corrected by the moon center coordinate (x center ,y center ) obtained by step 1) fitting. wherein, is the center pixel coordinate of the image image plane, K is the conversion matrix of the image physical imaging coordinate and the image coordinate system, and f is the equivalent focal length of the instrument. 36) Calculate the azimuth of the line joining the projected points of the two points A and B on the lunar surface in the image plane with respect to the image north direction from the results of step 35)

Images