Single corner reflector based sar satellite three-dimensional positioning error correction method and system

Abstract

The application discloses a kind of single corner reflection-based SAR satellite three-dimensional positioning error correction method and system, according to the space-time characteristics of multi-source error in spaceborne SAR positioning process, i.e. in a certain area range, its system error, propagation path error and earth model error have spatial correlation, and three direction coordinates in radar coordinate system are independent of each other, error calculation is carried out using the signal of single corner reflector, to realize simple and effective high-precision positioning of whole scene SAR data. The method does not depend on correcting system error, earth model and propagation path delay error one by one, establishes unified correction model using ground corner reflector, realizes three-dimensional high-precision positioning of spaceborne SAR image target. The method provides precise positioning coordinates of SAR image target, provides basis for interpretation of SAR observation results in geographical coordinate system and fusion with other geodetic survey data. The application effectively improves the three-dimensional positioning accuracy of target point in spaceborne SAR image.

Description

Technical Field

[0001] This invention belongs to the field of satellite remote sensing technology and relates to a method and system for correcting the three-dimensional precise positioning error of SAR satellite targets, specifically a method and system for correcting the three-dimensional positioning error of SAR satellites based on a single angular reflection. Background Technology

[0002] Spaceborne synthetic aperture radar (SAR) is an active microwave side-looking imaging radar carried on satellites. Because it has the ability to observe the Earth over a wide area, at all times, and in all weather conditions, it is an important means of acquiring information about the Earth's surface.

[0003] Spaceborne SAR imaging involves using sensors mounted on a satellite to collect backscattered signals from the observed area, transforming a three-dimensional observation scene into a two-dimensional image. Calculating the ground location from target points within the SAR image is the reverse process, seeking a solution from the two-dimensional image to three-dimensional space.

[0004] Currently, the process of solving for the 3D coordinates of spaceborne SAR target points mainly utilizes the range-Doppler positioning model. This model assumes that the SAR target satisfies range, Doppler, and Earth model requirements, relies on external elevation data for assistance, and is affected by various factors such as Earth model errors, imaging system errors, and propagation path delays. The resulting 3D positioning accuracy is relatively low, often ranging from a few meters to tens of meters. Existing positioning error correction methods obtain corrections by calculating geophysical and systematic error terms individually during the positioning process. However, this method lacks corrections for elevation deviations, making it difficult to obtain high-precision 3D positioning results for spaceborne SAR targets, especially failing to meet the fine-grained monitoring needs of urban scenarios. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for correcting SAR satellite three-dimensional positioning errors based on a single angular reflection.

[0006] The technical solution adopted by the method of the present invention is: a SAR satellite three-dimensional positioning error correction method based on single angular reflection, comprising the following steps:

[0007] Step 1: Acquire spaceborne SAR images and their imaging parameters and orbital data; the imaging parameters include the initial imaging azimuth time t. a,0 Initial imaging range at time t r,0 Pulse repetition frequency (PRF), distance sampling rate (RSF), satellite motion velocity (v) s / c And the microwave propagation speed v0;

[0008] Step 2: Reconstruct SAR image information in the frequency domain, detect target point T in the SAR image, and determine the target point's coordinates in the two-dimensional radar image, i.e., the sub-pixel coordinates in the row and column dimensions (l T ,p T );

[0009] Step 3: Use SAR image imaging parameter data to convert sub-pixel coordinate values ​​(l T ,p T ) Convert to two-dimensional radar geometric coordinates, that is, the coordinate values ​​of radar azimuth and range (a T ,r T );

[0010] Step 4: Obtain the radar elevation coordinates c of the target point T Obtain the three-dimensional radar coordinates of the target point (a T ,r T ,c T );

[0011] Step 5: Set the geodetic coordinates (X, Y, Z) of the corner reflector. CR ,Y CR H CR The radar coordinates of the target point are calculated inversely based on the SAR image imaging parameters and orbit data. CR ,r CR ,c CR ( ), compared with the estimated three-dimensional radar coordinates of the target point in the SAR image data Correspondingly, the difference between the radar coordinates retrieved from the corner reflector's geodetic coordinates and the estimated radar coordinates in the SAR image is used as the total error for correction, yielding correction parameters (Δa, Δr, Δc). These correction parameters are then applied to the estimated three-dimensional radar coordinates (a, Δr, Δc) of the target points in all SAR image data. T +Δa,r T +Δr,c T +Δc);

[0012] Step 6: Place (a) T +Δa,r T +Δr,c T +Δc) Input iteration distance - Doppler - Earth ellipsoid model to calculate the three-dimensional coordinates (X) of the target point in the geodetic coordinate system. T ,Y T H T ).

[0013] The technical solution adopted by the system of the present invention is: a SAR satellite three-dimensional positioning error correction system based on a single angular reflection, comprising the following modules:

[0014] Module 1 is used to acquire spaceborne SAR images and their imaging parameters and orbital data; the imaging parameters include the initial imaging azimuth time t.a,0 Initial imaging range at time t r,0 Pulse repetition frequency (PRF), distance sampling rate (RSF), satellite motion velocity (v) s / c And the microwave propagation speed v0;

[0015] Module 2 is used to reconstruct SAR image information in the frequency domain, detect target point T in the SAR image, and determine the target point's coordinates in the two-dimensional radar image, i.e., the sub-pixel coordinates of the row and column dimensions (l). T ,p T );

[0016] Module 3 is used to convert sub-pixel coordinates (l) from SAR image imaging parameter data. T ,p T ) Convert to two-dimensional radar geometric coordinates, that is, the coordinate values ​​of radar azimuth and range (a T ,r T );

[0017] Module 4 is used to obtain the radar elevation coordinates c of the target point. T Obtain the three-dimensional radar coordinates of the target point (a T ,r T ,c T );

[0018] Module 5 is used to input the geodetic coordinates (X, Y, Z) of the corner reflector. CR ,Y CR H CR The radar coordinates of the target point are calculated inversely based on the SAR image imaging parameters and orbit data. CR ,r CR ,c CR ( ), compared with the estimated three-dimensional radar coordinates of the target point in the SAR image data Correspondingly, the difference between the radar coordinates retrieved from the corner reflector's geodetic coordinates and the estimated radar coordinates in the SAR image is used as the total error for correction, yielding correction parameters (Δa, Δr, Δc). These correction parameters are then applied to the estimated three-dimensional radar coordinates (a, Δr, Δc) of the target points in all SAR image data. T +Δa,r T +Δr,c T +Δc);

[0019] Module 6 is used to transfer (a T +Δa,r T +Δr,c T +Δc) Input iteration distance - Doppler - Earth ellipsoid model to calculate the three-dimensional coordinates (X) of the target point in the geodetic coordinate system. T ,Y T H T ).

[0020] This invention utilizes the spatiotemporal characteristics of errors in spaceborne SAR positioning. Specifically, within a certain area, the systematic errors, atmospheric delay, and Earth model errors exhibit spatial correlation, while the coordinates in the three directions within the radar coordinate system are independent. Therefore, error correction is performed within the three-dimensional radar coordinate system, and error calculation is conducted using the signal from a single corner reflector within the SAR data acquisition range, thereby achieving simple and effective high-precision positioning of the entire SAR scene.

[0021] Advantages of this invention:

[0022] 1. This invention is applicable to the three-dimensional precise positioning correction of SAR data acquired by any satellite sensor.

[0023] 2. This invention only requires the coordinate position of one angular reflection as input to achieve high-precision positioning of targets in the entire SAR data scene, without the need to correct system errors, atmospheric delays and earth model errors one by one.

[0024] 3. This invention relies on data from only a single corner reflector, resulting in high accuracy and simple, quick calculation, making it easy to promote in engineering applications. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the method in an embodiment of the present invention. Detailed Implementation

[0026] To facilitate understanding and implementation of the present invention by those skilled in the art, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0027] Please see Figure 1 The present invention provides a SAR satellite three-dimensional positioning error correction method based on a single angular reflection, comprising the following steps:

[0028] Step 1: Acquire spaceborne SAR images and their imaging parameters and orbital data; imaging parameters include initial imaging azimuth and time t. a,0 Initial imaging range at time t r,0 Pulse repetition frequency (PRF), distance sampling rate (RSF), satellite motion velocity (v) s / c And the microwave propagation speed v0;

[0029] Step 2: Reconstruct SAR image information in the frequency domain, detect the target point location in the SAR image, and determine the target point's coordinates in the two-dimensional radar image, i.e., the sub-pixel coordinate values ​​in the row and column dimensions (l T ,p T );

[0030] In this embodiment, the sub-pixel coordinate values ​​are determined by using fast frequency domain interpolation to reconstruct SAR slice image information; this method only requires inputting 7×7 pixels within the target point's neighborhood, f={f(x1),…f(x2),…f(x3),…f(x4),…f(x5),…f(x6),…f(x7 ... i ),…f(x n )},x i =(l i ,p i (i = 1:7), the signal is transformed into the frequency domain through Discrete Fourier Transform (DFT), and zero-padding is performed in the frequency domain to achieve fast signal interpolation. Then, the signal is transformed into the spatial domain through Inverse Fourier Transform (IDFT) to obtain a 512×512 pixel SAR small area image reconstructed in the spatial domain. The position of the target is determined by searching for the position of the strongest signal in the reconstructed slice image, and then the pixel position of the SAR image coordinates is calculated back to obtain the sub-pixel floating-point coordinate value of the target.

[0031] Step 3: Use SAR image imaging parameter data to convert sub-pixel coordinate values ​​(l T ,p T ) Convert to two-dimensional radar geometric coordinates, that is, the coordinate values ​​of radar azimuth and range (a T ,r T );

[0032] In this embodiment, the formula for transforming radar image coordinates to radar geometric coordinates is as follows:

[0033]

[0034]

[0035] Since the determination of the azimuth position is related to the motion of the satellite platform, traditional methods use approximate calculations. This embodiment uses the satellite platform's motion speed v... s / c Projected trajectory velocity v on the ground g / t Furthermore, the accuracy of azimuth geometric coordinate calculation is improved by employing time-dimensional integration.

[0036] Step 4: Obtain the radar elevation coordinates c of the target point T Obtain the three-dimensional radar coordinates of the target point (a T ,r T ,c T );

[0037] In this embodiment,

[0038]

[0039] n SAR image data constitute a multi-baseline interferometric SAR image dataset, forming an interferometric phase matrix. Baseline matrix Sum distance matrix λ is the sensor's operating wavelength; the radar elevation coordinates of the target point are obtained based on least squares estimation using phase quality weighting.

[0040]

[0041] The baseline matrix and the distance matrix constitute the diagonal coefficient matrix for solving the problem. The cofactor matrix is ​​a diagonal matrix in It is a quality characterization of each phase observation; the residual phase quantity of the observed phase is introduced. Where ΔD is the nonlinear motion of the target, and Δα atmos It is the residual amount of the atmosphere. It is thermal noise. The residual components of each observation phase are determined by estimating the linear deformation phase contribution, the atmospheric phase contribution from spatiotemporal filtering, and the topographic-related phase contribution within the observation phase, and the standard deviation of the residual phase is obtained statistically.

[0042] Step 5: Set the geodetic coordinates (X, Y, Z) of the corner reflector. CR ,Y CR H CR The radar coordinates of the target point are calculated inversely based on the SAR image imaging parameters and orbit data. CR ,r CR ,c CR ( ), compared with the estimated three-dimensional radar coordinates of the target point in the SAR image data Correspondingly, the difference between the radar coordinates retrieved from the corner reflector's geodetic coordinates and the estimated radar coordinates in the SAR image is used as the total error for correction, yielding correction parameters (Δa, Δr, Δc). These correction parameters are then applied to the estimated three-dimensional radar coordinates (a, Δr, Δc) of the target points in all SAR image data. T +Δa,r T +Δr,c T +Δc);

[0043] In this embodiment, the positional deviation provided by a single corner reflector is introduced as a global error correction. SAR image target location is affected by various errors, including sensor timing errors, azimuth system offsets, tropospheric and ionospheric delays, crustal movement, Earth solid tides, tidal load effects, polar motion, atmospheric tidal loads, and reference point elevation errors. In a single SAR image, these errors exhibit systematic or spatial correlations and are independent within the radar coordinate system (range, azimuth, and elevation directions). The target position (a) can be corrected by solving for the deviations (Δa, Δr, Δc) in the three directions. T +Δa,r T +Δr,cT +Δc);

[0044]

[0045]

[0046]

[0047] Step 6: Place (a) T +Δa,r T +Δr,c T +Δc) Input iteration distance - Doppler - Earth ellipsoid model to calculate the three-dimensional coordinates (X) of the target point in the geodetic coordinate system. T ,Y T H T ).

[0048] In this embodiment, the relationship between the target radar elevation and the geodetic height is established:

[0049] H(c T ) = c T sinθ T ;

[0050] The radar three-dimensional coordinate position of the target (a T +Δa,r T +Δr,c T +Δc) is directly substituted into the precise positioning range-Doppler-Ellipsoid (RDE) model, combined with the satellite orbit vector data S(t a ) and satellite motion data V(t a The three-dimensional position T = (X_i, X_j) of the target in the geodetic coordinate system is solved through multiple iterations. T ,Y T Z T The iteration terminates when the coordinate difference between the two solutions converges.

[0051] Distance function:

[0052] Doppler function:

[0053] Earth ellipsoid function:

[0054] Wherein, S(t) a ) is the satellite at zero Doppler time t a orbital position, f D (t a ) is at zero Doppler time t a Doppler centroid value, V(t) a) is the satellite at zero Doppler time t a The velocity of motion, m and f are the major and minor semi-axises of the Earth's ellipsoid, H(c T ) is using c T Calculated geodetic height.

[0055] The following uses publicly available Sentinel-1 satellite SAR image processing as an example to illustrate the technical implementation scheme of the present invention, but it should not be construed as a limitation on the technical scheme.

[0056] Step 1. Acquire multiple SAR images and perform data reading, image registration, baseline estimation, and interferometry. The combination of interferometric pairs can fully consider baseline conditions.

[0057] Step 2. Extract the subpixel coordinates of the target point in the SAR image at the image level. Perform fast frequency domain interpolation in the neighborhood of the target point and complete the zero-padding in the frequency domain by Fourier transform. Select the position with the strongest intensity in the local area, which is the subpixel image coordinate of the target point.

[0058] Step 3. Based on the image imaging parameters, convert the image coordinates to geometric coordinates, which are now two-dimensional radar geometric coordinates, namely azimuth and range coordinates. From the phase observations of the interferometric pair, obtain the radar elevation coordinates of the target point using least squares estimation based on phase quality weighting. The azimuth, range, and elevation coordinates describe the target's three-dimensional spatial position in the radar coordinate system.

[0059] Step 4. The geodetic coordinates of a single corner reflector within the SAR image acquisition range are calculated using imaging geometry to determine its three-dimensional position in the radar coordinate system. This position is then combined with the three-dimensional coordinates calculated from the spaceborne SAR image and incorporated into the error model. Due to the influence of various errors such as timing error, azimuth system offset, tropospheric and ionospheric delay, crustal movement, Earth solid tides, ocean tidal load effects, polar motion, atmospheric tidal load, and reference point elevation error, the errors exhibit systematicity and spatial correlation within the entire image. This yields the global corrections for the target point in the entire SAR image in three directions within the radar coordinate system.

[0060] Step 5. Substitute the corrected radar 3D coordinates into the range-Doppler-Earth ellipsoid model to calculate the geodetic coordinates of the target point, and complete the 3D precise positioning.

[0061] This invention does not rely on individually correcting systematic errors, geophysical errors, and propagation path delay errors. Instead, it utilizes a ground corner reflector to establish a unified correction model, achieving high-precision three-dimensional positioning of targets in spaceborne SAR images. This effectively improves the three-dimensional positioning accuracy of target points in spaceborne SAR images. This invention provides precise positioning coordinates for SAR image targets, laying the foundation for the interpretation of SAR Earth observation results in geographic coordinate systems and their fusion with other geodetic data.

[0062] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A method for correcting SAR satellite three-dimensional positioning errors based on a single angular reflection, characterized in that, Includes the following steps: Step 1: Acquire spaceborne SAR images and their imaging parameters and orbital data; the imaging parameters include the initial imaging azimuth time t. a，0 Initial imaging range at time t r，0 Pulse repetition frequency (PRF), distance sampling rate (RSF), satellite motion velocity (v) s / c And the microwave propagation speed v0; Step 2: Reconstruct SAR image information in the frequency domain, detect the position of target point T in the SAR image, and determine the target point's coordinates in the two-dimensional radar image, i.e., the sub-pixel coordinates of the row and column dimensions (l T p T ); Step 3: Use SAR image imaging parameter data to convert sub-pixel coordinate values ​​(l T p T ) Convert to two-dimensional radar geometric coordinates, that is, the coordinate values ​​of radar azimuth and range (a T r T ); Step 4: Obtain the radar elevation coordinates of the target point to obtain the three-dimensional radar coordinates of the target point (a T r T c T ); Step 5: Set the geodetic coordinates (X, Y, Z) of the corner reflector. CR Y CR H CR The radar coordinates of the target point are calculated inversely based on the SAR image imaging parameters and orbit data. CR r CR c CR ( ), compared with the estimated three-dimensional radar coordinates of the target point in the SAR image data Correspondingly, the difference between the radar coordinates retrieved from the corner reflector's geodetic coordinates and the estimated radar coordinates in the SAR image is used as the total error for correction, obtaining correction parameters (Δa, Δr, Δc). These correction parameters are then applied to the estimated three-dimensional radar coordinates (a, Δr, Δc) of the target points in all SAR image data. T +Δa, r T +Δr, c T +Δc); Step 6: Place (a) T +Δa, r T +Δr, c T +Δc) Input iteration distance - Calculate the three-dimensional coordinates (X, Y, F, Z) of the target point in the geodetic coordinate system using the Doppler-Earth ellipsoid model. T Y T H T ).

2. The SAR satellite three-dimensional positioning error correction method based on a single angular reflection according to claim 1, characterized in that: In step 2, the sub-pixel coordinates are determined using fast frequency domain interpolation to reconstruct SAR slice image information; the input target point's neighborhood contains 7×7 pixels f = {f(x1), ..., f(x2)}. i ), …f(x) n )},x i =(l i p i (i = 1:7), the signal is converted to the frequency domain through discrete Fourier transform, and zero-padding is performed in the frequency domain to achieve fast signal interpolation. Then, the signal is converted to the spatial domain through inverse Fourier transform to obtain a 512×512 pixel SAR slice image reconstructed in the spatial domain. The position of the target is determined by searching for the strongest signal position in the reconstructed small area image, and then the pixel position of the SAR image coordinates is calculated back to obtain the sub-pixel floating-point coordinate value of the target.

3. The SAR satellite three-dimensional positioning error correction method based on a single angular reflection according to claim 1, characterized in that: In step 3, during the transformation from radar image coordinates to radar geometric coordinates, the range coordinates are calculated as follows: For the azimuth coordinates, the satellite platform's motion speed v s / c Projected trajectory velocity v on the ground g / t Furthermore, the accuracy of azimuth geometric coordinate calculation is improved by employing time-dimensional integration.

4. The SAR satellite three-dimensional positioning error correction method based on a single angular reflection according to claim 1, characterized in that: In step 4, n SAR image data constitute a multi-baseline interferometric SAR image dataset, forming an interferometric phase matrix. Baseline matrix Sum distance matrix λ is the operating wavelength of the sensor; Radar elevation coordinates of target points are obtained based on least squares estimation with phase quality weighting. The baseline matrix and the distance matrix constitute the diagonal coefficient matrix for solving the problem. The cofactor matrix is ​​a diagonal matrix in It is a quality characterization of each phase observation; the residual phase quantity of the observed phase is introduced. Where ΔD is the nonlinear motion of the target, and Δα atmos It is the residual amount of the atmosphere. It is thermal noise. The residual components of each observation phase are determined by estimating the linear deformation phase contribution, the atmospheric phase contribution from spatiotemporal filtering, and the topographic-related phase contribution within the observation phase, and the standard deviation of the residual phase is obtained statistically.

5. The SAR satellite three-dimensional positioning error correction method based on a single angular reflection according to any one of claims 1, characterized in that: In step 5, the deviations (Δa, Δr, Δc) in three directions are calculated to correct the target position (a). T +Δa, r T +Δr, c T +Δc); 6. The SAR satellite three-dimensional positioning error correction method based on a single angular reflection according to claim 1, characterized in that: In step 6, the relationship between the target radar elevation and the geodetic height is established: H(c T )=c T sinθ T ; The radar three-dimensional coordinate position of the target (a T +Δa, r T +Δr, c T +Δc) is directly substituted into the precise positioning distance-Doppler-Earth ellipsoid model, combined with the satellite orbit vector data S(t a ) and satellite motion data V(t a The three-dimensional position T = (X_i, X_j) of the target in the geodetic coordinate system is solved through multiple iterations. T Y T H T The iteration terminates when the coordinate difference between the two solutions converges.

7. A SAR satellite three-dimensional positioning error correction system based on a single angular reflection, characterized in that, Includes the following modules: Module 1 is used to acquire spaceborne SAR images and their imaging parameters and orbital data; the imaging parameters include the initial imaging azimuth time t. a，0 Initial imaging range at time t r，0 Pulse repetition frequency (PRF), distance sampling rate (RSF), satellite motion velocity (v) s / c And the microwave propagation speed v0; Module 2 is used to reconstruct SAR image information in the frequency domain, detect target point T in the SAR image, and determine the target point's coordinates in the two-dimensional radar image, i.e., the sub-pixel coordinate values ​​in the row and column two-dimensional plane (l T p T ); Module 3 is used to convert sub-pixel coordinates (l) from SAR image imaging parameter data. T p T ) Convert to two-dimensional radar geometric coordinates, that is, the coordinate values ​​of radar azimuth and range (a T r T ); Module 4 is used to obtain the radar elevation coordinates c of the target point. T Obtain the three-dimensional radar coordinates of the target point (a T r T c T ); Module 5 is used to input the geodetic coordinates (X, Y, Z) of the corner reflector. CR Y CR H CR The radar coordinates of the target point are calculated inversely based on the SAR image imaging parameters and orbit data. CR r CR c CR ( ), compared with the estimated three-dimensional radar coordinates of the target point in the SAR image data Correspondingly, the difference between the radar coordinates retrieved from the corner reflector's geodetic coordinates and the estimated radar coordinates in the SAR image is used as the total error for correction, obtaining correction parameters (Δa, Δr, Δc). These correction parameters are then applied to the estimated three-dimensional radar coordinates (a, Δr, Δc) of the target points in all SAR image data. T +Δa, r T +Δr, c T +Δc); Module 6 is used to transfer (a T +Δa, r T +Δr, c T +Δc) Input iteration distance - Doppler - Earth ellipsoid model to calculate the three-dimensional coordinates (X) of the target point in the geodetic coordinate system. T Y T H T ).

Images