A three-dimensional imaging method based on squint mode SAR tomographic data

Abstract

The three-dimensional imaging method based on the squint mode SAR tomography data of the present invention obtains the precise ephemeris of N re-orbit satellites, obtains the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of each re-orbit satellite trajectory according to the precise ephemeris of the re-orbit satellite; calculates the re-orbit satellite trajectory k according to the scene center position and the re-orbit satellite position of the reference trajectory k0 i The geometric correlation coefficient of the image pair with the reference trajectory k0; according to the heavy orbit satellite trajectory k i The accurate ephemeris of the image is used to calculate the effective vertical baseline of the re-orbit aperture center to the reference trajectory k0; the re-orbit satellite trajectory k is calculated based on the geometric correlation coefficient of the image pair and the effective vertical baseline. i The height inversion accuracy CRLB of the selected heavy orbit satellite trajectory k i The optimal position of the elevation inversion accuracy CRLB is used as the trajectory of the re-orbit satellite k i The center of the heavy orbit aperture is used to collect the heavy orbit satellite trajectory k i SAR image; repeat the above steps until all heavy track aperture centers are optimized; use compressed sensing method to estimate target elevation, and obtain the three-dimensional position of the target in the geodetic coordinate system according to the elevation direction of the squint mode to improve the accuracy of SAR tomographic three-dimensional imaging.

Description

Technical Field

[0001] The invention belongs to the technical field of synthetic aperture radar, and in particular relates to a three-dimensional imaging method based on squint mode SAR tomography data. Background Art

[0002] Synthetic Aperture Radar (SAR) tomography uses multiple SAR images with different observation angles to form a height synthetic aperture, which can separate overlapping targets in the same resolution unit to achieve three-dimensional imaging. Estimating the height position of overlapping targets is one of the most important purposes of SAR tomography.

[0003] With the increase in the altitude of spaceborne SAR and the diversification of working modes, the squint mode is becoming more and more common in spaceborne SAR. In medium-orbit (orbit altitude above 1500km and below geosynchronous orbit) and high-orbit SAR (orbit in geosynchronous orbit), squint is a very necessary means to achieve wide bandwidth. In particular, for high-orbit SAR, squint is conducive to achieving large-scale regional coverage.

[0004] In the application of tomographic SAR, airborne SAR and low-orbit SAR (orbit altitude less than 1500km) usually consider the normal and side-view working modes, and their beams usually point to the zero Doppler plane. Therefore, by taking the zero Doppler center (ZDC) moment as the aperture center moment of SAR imaging, the image sequence with the largest correlation coefficient can be acquired. However, for the squint mode SAR, the ZDC method is difficult to guarantee the maximum correlation coefficient. In addition, in the squint mode, the accuracy of the traditional three-dimensional imaging method decreases due to the difference in geometric model from the normal and side-view modes. In order to achieve the optimal tomographic performance of squint mode SAR, it is necessary to select the optimal aperture center moment in combination with the squint observation geometry, and perform three-dimensional imaging according to the squint mode, which has not been mentioned in the existing SAR tomography research. Summary of the invention

[0005] The present invention overcomes one of the deficiencies of the prior art and provides a three-dimensional imaging method based on squint mode SAR tomography data, which can select the best aperture center moment for each track according to the SAR squint geometry, thereby improving the three-dimensional imaging accuracy of SAR tomography.

[0006] According to one aspect of the present invention, the present invention provides a three-dimensional imaging method based on squint mode SAR tomography data, the method comprising:

[0007] Obtain N re-orbit satellite precise ephemerides, and obtain the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of each re-orbit satellite trajectory according to the re-orbit satellite precise ephemerides, where N is a positive integer;

[0008] Select any one of the heavy orbit satellite trajectories as the reference trajectory k0;

[0009] The re-orbit satellite trajectory k is calculated based on the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of the reference trajectory k0. i The geometric correlation coefficient of the image pair with the reference trajectory k0, where i=1, 2, ..., N-1;

[0010] According to the re-orbit satellite trajectory k i The precise ephemeris of the satellite is used to calculate the trajectory k of the re-orbit satellite. i The effective vertical baseline to the center of the re-track aperture of the reference trajectory k0;

[0011] According to the re-orbit satellite trajectory k i The geometric correlation coefficient of the image pair with the reference trajectory k0 and the effective vertical baseline are used to calculate the re-orbit satellite trajectory k i CRLB of elevation inversion accuracy;

[0012] Select the heavy orbit satellite trajectory k i The optimal position of the CRLB with the height inversion accuracy is taken as the re-orbit satellite trajectory k i The center of the re-orbit aperture is used to collect the re-orbit satellite trajectory k i SAR images;

[0013] Repeat the above steps until all heavy orbit satellite tracks k i The heavy rail aperture center optimization is completed;

[0014] The compressed sensing method is used to estimate the target elevation, and the three-dimensional position of the target in the geodetic coordinate system is obtained according to the elevation direction of the squint mode.

[0015] In a possible implementation, the re-orbit satellite trajectory k is calculated based on the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of the reference trajectory k0. i The geometric correlation coefficient of the image pair with the reference trajectory k0 includes:

[0016] In the re-orbit satellite trajectory k i On the search, the scene center position at the moment of the heavy track aperture center of the reference trajectory k0 is found Recent Locations

[0017] At the stated location As the center, select the satellite trajectory intervals of 2L before and after, and evenly divide the satellite trajectory interval into M points;

[0018] For the mth point, according to the re-orbit satellite trajectory k i The precise ephemeris is used to obtain the velocity of the point m The unit vector of the line of sight direction is calculated according to the position of the re-orbit satellite at the center of the re-orbit aperture of the reference trajectory k0 Calculate the unit vector The projection matrix for the normal vector

[0019] According to the projection matrix P k0 Calculate the ground normal vector And calculate the ground normal vector The projection matrix for the normal vector

[0020] According to the unit vector Projection Matrix P k0 , the ground normal vector The projection matrix Pg calculates the trajectory k of the re-orbit satellite i The geometric correlation coefficient γ of the image pair with the reference trajectory k0.

[0021] In one possible implementation,

[0022] The geometric correlation coefficient

[0023] Where c is the speed of light, λ is the wavelength of the radar payload, and B is the bandwidth of the transmitted signal. represents the slant distance from the center of the heavy track aperture of the reference trajectory k0 to the center of the scene;

[0024] Represents the reference trajectory k0 and the current re-orbit satellite trajectory k i The difference between the sight unit vector from the current position to the center of the scene, express The projection on the ground,

[0025] represents the projection of the synthetic aperture vector of the reference trajectory k0 on the ground;

[0026] They respectively represent the ground equidistant direction and the equal Doppler direction of the reference trajectory k0 imaging.

[0027] In a possible implementation, the re-orbit satellite trajectory k i The precise ephemeris of the satellite is used to calculate the trajectory k of the re-orbit satellite. i The effective vertical baseline to the center of the re-track aperture of the reference trajectory k0 includes:

[0028] Calculate the current heavy orbit satellite trajectory k i The position of the re-orbit satellite at the time of the re-orbit aperture center The re-orbit satellite position at the time of the re-orbit aperture center of the reference trajectory k0 Baseline

[0029] The effective vertical baseline

[0030] In a possible implementation, the elevation inversion accuracy is Where γ is the trajectory k of the re-orbit satellite i The geometric correlation coefficient of the image pair with the reference trajectory k0, λ is the radar payload wavelength, b n is the trajectory k of the heavy orbit satellite i The effective vertical baseline at the time instant of the re-track aperture center to the reference trajectory k0.

[0031] The three-dimensional imaging method based on the squint mode SAR tomography data of the present invention obtains N accurate ephemeris of repeated orbit satellites, obtains the scene center position and the repeated orbit satellite position at the repeated orbit aperture center moment of each repeated orbit satellite trajectory according to the accurate ephemeris of the repeated orbit satellite, N is a positive integer; selects any one of the repeated orbit satellite trajectories as a reference trajectory k0; calculates the repeated orbit satellite trajectory k0 according to the scene center position and the repeated orbit satellite position at the repeated orbit aperture center moment of the reference trajectory k0 i The geometric correlation coefficient of the image pair with the reference trajectory k0, where i=1, 2, ..., N-1; according to the re-orbit satellite trajectory k i The precise ephemeris of the satellite is used to calculate the trajectory k of the re-orbit satellite. i to the effective vertical baseline at the center of the re-orbit aperture of the reference trajectory k0; according to the re-orbit satellite trajectory k i The geometric correlation coefficient of the image pair with the reference trajectory k0 and the effective vertical baseline are used to calculate the re-orbit satellite trajectory k i CRLB of the height inversion accuracy; Select the heavy orbit satellite trajectory k i The optimal position of the CRLB with the height inversion accuracy is taken as the re-orbit satellite trajectory k i The center of the re-orbit aperture is used to collect the re-orbit satellite trajectory k i Repeat the above steps until all the heavy orbit satellite tracks k i The center of the heavy track aperture is optimized; the target elevation is estimated using the compressed sensing method, and the three-dimensional position of the target in the geodetic coordinate system is obtained according to the elevation direction of the squint mode. The geometric correlation coefficients of each position on the trajectory and the main track are accurately predicted through SAR observation geometry, and the orbit position with the best theoretical elevation inversion accuracy is obtained, the optimal tomographic data acquisition is completed, and the accuracy of tomographic three-dimensional imaging is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings are used to provide a further understanding of the technical solution of the present application or the prior art, and constitute a part of the specification. Among them, the accompanying drawings expressing the embodiments of the present application are used together with the embodiments of the present application to explain the technical solution of the present application, but do not constitute a limitation on the technical solution of the present application.

[0033] Figure 1 A flow chart of a three-dimensional imaging method based on squint mode SAR tomography data according to an embodiment of the present disclosure is shown;

[0034] Figure 2 A flow chart of a three-dimensional imaging method based on squint mode SAR tomography data according to another embodiment of the present disclosure is shown;

[0035] Figure 3 A schematic diagram showing the principle of acquiring data based on SAR tomography in a squint mode according to an embodiment of the present disclosure is shown;

[0036] Figure 4 A schematic diagram of the geometric observation of SAR tomography in a squint mode according to an embodiment of the present disclosure is shown;

[0037] Figure 5 A schematic diagram of a squint mode SAR tomography simulation scene according to an embodiment of the present disclosure is shown;

[0038] Figure 6 A schematic diagram of a vertical baseline of an oblique viewing mode SAR tomography simulation according to an embodiment of the present disclosure is shown;

[0039] Figure 7 A schematic diagram showing a detection result of the number of overlaps according to an embodiment of the present disclosure is shown;

[0040] Figure 8 A schematic diagram of a three-dimensional imaging result based on a traditional mode according to an embodiment of the present disclosure is shown;

[0041] Fig. 9 A schematic diagram of a three-dimensional imaging result based on squint mode SAR tomography data according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0042] The following will describe the implementation methods of the present invention in detail with reference to the accompanying drawings and embodiments, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve corresponding technical effects can be fully understood and implemented accordingly. The embodiments of the present application and the various features in the embodiments can be combined with each other without conflict, and the technical solutions formed are all within the protection scope of the present invention.

[0043] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer such as a set of computer executable instructions. Also, although a logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a sequence different from that here.

[0044] Figure 1 and Figure 2 1 and 2 show a flow chart of a three-dimensional imaging method based on squint mode SAR tomography data according to an embodiment of the present disclosure. Figure 2 As shown in Figure 1, this method requires multiple SAR images to realize the squint mode SAR tomography. The satellite can revisit the same area or use an array to obtain multiple images simultaneously, that is, the satellite illuminates the same area multiple times, such as Figure 3 shown.

[0045] Specific as Figure 1 As shown, the method may include:

[0046] Step S1: Obtain N repeated-orbit satellite precise ephemerides, and obtain the scene center position and the repeated-orbit satellite position at the repeated-orbit aperture center moment of each repeated-orbit satellite trajectory according to the repeated-orbit satellite precise ephemerides, where N is a positive integer.

[0047] Step S2: Select any one of the heavy-orbit satellite trajectories as the reference trajectory k0.

[0048] Considering that SAR tomography uses multiple re-orbit data, it is necessary to obtain the precise satellite ephemeris of each re-orbit satellite passing over the same area for a period of time. If N satellite orbits are obtained, N is a positive integer. The satellite trajectory of each re-orbit (the scene center position and the re-orbit satellite position at the re-orbit aperture center moment) is calculated based on the precise satellite ephemeris.

[0049] SAR tomography usually uses a certain trajectory as a reference orbit, sets the reference orbit (specified by the user), and sets its number as k0. The user specifies the SAR working mode as oblique viewing mode or downward viewing mode. According to the SAR working mode, the average synthetic aperture length of each target in the scene is calculated, which can be completed according to the radar beam scanning strategy and the SAR geometric relationship, which will not be described here. Satellite motion not only has size but also direction, and the synthetic aperture can be represented by a vector, that is, is called the synthetic aperture vector.

[0050] Step S3: Calculate the re-orbit satellite trajectory k according to the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of the reference trajectory k0 i The geometric correlation coefficient of the image pair with the reference trajectory k0, where i=1, 2, ..., N-1.

[0051] The SAR data acquisition strategy of the present invention is to collect data for all heavy orbit satellite tracks (orbits) other than the reference orbit.

[0052] In one example, the steps may include:

[0053] Step S31: In the re-orbit satellite trajectory k i On the search, the scene center position at the moment of the heavy track aperture center of the reference trajectory k0 is found Recent Locations

[0054] Step S32: Using the position As the center, select the satellite trajectory intervals of 2L before and after, and evenly divide the satellite trajectory interval into M points;

[0055] Step S33: For the mth point, according to the re-orbit satellite trajectory k i The precise ephemeris of the point m is used to obtain the velocity of the point m. The unit vector of the line of sight direction is calculated according to the position of the re-orbit satellite at the center of the re-orbit aperture of the reference trajectory k0 Calculate the unit vector The projection matrix for the normal vector

[0056] Step S34: According to the projection matrix P k0 Calculate the ground normal vector And calculate the ground normal vector The projection matrix for the normal vector

[0057] Step S35: According to the unit vector Projection Matrix P k0 , the ground normal vector The projection matrix Pg calculates the trajectory k of the re-orbit satellite i The geometric correlation coefficient γ of the image pair with the reference trajectory k0.

[0058] For example, the position of the re-orbit satellite at the time of the re-orbit aperture center of the reference orbit k0 is (The origin of each satellite position vector is the center of the earth), and the corresponding scene center position at the time of the re-orbit aperture center is Select a heavy rail track, denoted as k i , and k≠k0, such as Figure 2 As shown. i On the track, the search is Recent Locations And thought Center, select the 2L trajectory intervals before and after, and divide the trajectory interval evenly into M points.

[0059] For the mth point (m=1, 2, ..., M), obtain the velocity of the point according to the precise ephemeris Satellite position at the reference orbit aperture center Calculate the unit vector of the view direction And further calculate The projection matrix for the normal vector According to P k0 Calculate the ground normal vector And further calculate The projection matrix for the normal vector

[0060] Calculate the theoretical geometric correlation coefficient γ between the current position of the current orbit and the momentary position of the reference orbit aperture center. The calculation method is:

[0061]

[0062] Where c is the speed of light, λ is the wavelength of the radar payload, and B is the bandwidth of the transmitted signal. represents the slant distance from the center of the heavy track aperture of the reference trajectory k0 to the center of the scene;

[0063] Represents the reference trajectory k0 and the current re-orbit satellite trajectory k i The difference between the sight unit vector from the current position to the center of the scene, express The projection on the ground, represents the projection of the synthetic aperture vector of the reference trajectory k0 on the ground;

[0064] They represent the ground equidistance direction and the equi-Doppler direction of the reference trajectory k0 imaging. Some vector diagrams are shown in Figure 4 Schematic diagram of the squint mode SAR tomographic observation geometry shown.

[0065] Formula (1) represents the geometric correlation coefficient of any squint mode SAR image pair, and its derivation process is as follows:

[0066] Generally, the envelope difference in the range caused by the change in the sight line direction of the two observations is very small, so the envelope difference caused by the sight line difference between the two imaging can be ignored. However, the target delay phase caused by the sight line difference between the two imaging cannot be ignored, so after the kth i After acquiring echoes and imaging in each track, the center of the scene The point target expansion function [1] of the target at

[0067]

[0068] in is the position of any point on the ground, is the complex scattering coefficient of the target, A k represents the envelope of the point target spread function. Assuming that there are multiple scattering points on the ground, The imaging result at

[0069]

[0070] Consider a uniform ground surface, i.e. The phases in obey independent uniform distributions, then

[0071]

[0072] where σ 0 is the average radar cross-section area of ​​the ground. It is not general, so we take k = 1 and k = 2 to examine the cross-correlation of the two images. According to (4), we can get The cross-correlation of the pixels is:

[0073]

[0074] in They are the orbital positions corresponding to the imaging aperture center moments of the 1st and 2nd tracks respectively.

[0075] Considering the envelope Usually about The center is symmetric, so here the difference in the slope distance in the exponential term in equation (5) is by First, we perform a local linear expansion of the slant distance, that is,

[0076]

[0077] Similarly, we can get:

[0078] remember This formula represents the space with The neighborhood centered at is . The slope difference is Then there is

[0079]

[0080] in, It represents the change of the line of sight of the SAR platform between two observations.

[0081] make The cross-correlation can be expressed as:

[0082]

[0083] The last equal sign is true because is located in the ground plane, so And the projection matrix is ​​equal to its own transpose. Formula (8) is the two ground gradient directions along the ground distance direction and The surface integral of the product of the squares of the two sinc(·) functions, however, in the squint mode, the ground distance and the projection of the equivalent synthetic aperture on the ground Usually they are not orthogonal, which makes the envelope terms in the integrated terms coupled with each other in the two-dimensional plane. Therefore, it is necessary to find a suitable calculation method.

[0084] To calculate the above integral, we consider using the integral substitution method to find a non-orthogonal basis to decouple the integral. We decompose the ground into and Orthogonal direction (i.e. equidistantly) and Orthogonal direction Right now

[0085]

[0086] Then any Can be decomposed into the following forms:

[0087]

[0088] Notice therefore

[0089]

[0090]

[0091] Note that although and They are not orthogonal, but the result of this basis selection method is that the integral result produces a constant multiple. This multiple is related to the determinant of the Jacobian matrix corresponding to the basis transformation, but because the correlation coefficient is the normalized cross-correlation amplitude, this constant multiple does not affect the correlation coefficient. So the cross-correlation can be further simplified as:

[0092]

[0093] Finally, according to sinc 2 (·) The properties of Fourier transform are obtained

[0094]

[0095] When the two tracks completely overlap, the line of sight is At this time, the cross-correlation is the power of the image. The correlation coefficient is defined as the normalized cross-correlation. Therefore, the correlation coefficient of two images can be expressed as

[0096]

[0097] Step S4: According to the heavy orbit satellite trajectory k i The precise ephemeris of the satellite is used to calculate the trajectory k of the re-orbit satellite. i The effective vertical baseline at the time instant of the re-track aperture center to the reference trajectory k0.

[0098] Step S5: According to the heavy orbit satellite trajectory k i The geometric correlation coefficient and effective vertical baseline of the image pair with the reference trajectory k0 are used to calculate the re-orbit satellite trajectory k i CRLB of elevation inversion accuracy;

[0099] Step S6: Select the heavy orbit satellite trajectory k i The optimal position of the CRLB with the height inversion accuracy is taken as the re-orbit satellite trajectory k i The center of the heavy orbit aperture is used to collect the heavy orbit satellite trajectory k i SAR image.

[0100] For example, to calculate the current orbital position The moment position of the reference orbit aperture center Baseline Then calculate the effective vertical baseline b n , calculated as:

[0101]

[0102] The purpose of calculating the vertical baseline is to obtain the relationship between the interference phase between images and the target elevation. The derivation process of formula (15) is as follows:

[0103] Considering that all images will be registered with the image obtained by the reference track (called the main image) during the SAR tomography process, each target will be registered to the same resolution unit. Since SAR imaging processing is the process of obtaining the imaging results of the target in a certain plane (usually the ground plane), when the target has a certain elevation (that is, not on the ground plane), it will be projected to the ground plane along the elevation direction after imaging. Since each image is registered based on the main image, the elevation direction is the intersection of the reference track equidistance plane and the equal Doppler plane, which can be expressed as

[0104]

[0105] Usually after image registration, each image needs to be de-skewed, that is, the phase introduced by the flat ground is removed from the phase of each image. The slant distance of the first image after de-skew can be expressed as:

[0106]

[0107] in represents the target elevation. Similarly, the slant distance of the second image after deslanting can be expressed as Therefore, the slant distance corresponding to the interference phase of the two images is

[0108] Considering that the difference in sight lines between observations is usually small,

[0109]

[0110] From formula (18), we can obtain the equivalent vertical baseline, namely That is formula (19).

[0111] After obtaining the vertical baseline, the CRLB of the elevation inversion accuracy corresponding to the current orbit position is calculated according to the theoretical geometric correlation coefficient of each image pair. The calculation method is:

[0112]

[0113] Step S7: Repeat the above steps until all the heavy orbit satellite trajectories k i The heavy rail aperture center optimization is completed.

[0114] Step S8: Use the compressed sensing method to estimate the target elevation, and obtain the three-dimensional position of the target in the geodetic coordinate system according to the elevation direction of the squint mode.

[0115] The target elevation is estimated by using compressed sensing combined with generalized likelihood ratio detection. For a certain pixel in the main image, before three-dimensional imaging, it is usually assumed that the target is located on the ground plane. Assume that its coordinates in the geodetic coordinate system are Assuming that the estimated target elevation is h′, the actual three-dimensional position of the target in the geodetic coordinate system is:

[0116]

[0117] At this point, all steps are completed.

[0118] Application examples:

[0119] Figure 5 and Figure 6 The schematic diagram of the squint mode SAR tomography simulation scene and the schematic diagram of the simulation vertical baseline are shown respectively; Figure 7 A schematic diagram showing the detection result of the number of overlaps is shown; Figure 8A schematic diagram of three-dimensional imaging results based on the traditional mode is shown; Fig. 9 A schematic diagram of three-dimensional imaging results based on squint mode SAR tomography data is shown.

[0120] In this example, the satellite orbit altitude is 600km and the incidence angle is 45°. The type works in the oblique viewing mode, and the aperture center has a 250km along-track baseline compared to the straight side view case. The radar payload parameters are shown in Table 1. The ground range resolution is approximately 1m and the azimuth resolution is 1m. We simulated a ground plane in the center of the scene, on which a building with a length of 100m, a width of 30m and a roof height of 100m was placed, as shown in Table 1. Figure 5 As shown in the figure. The wall of the building is parallel to the satellite trajectory. Affected by the satellite's direction, the sides and back of the building are actually not observable, so we did not set up the corresponding plane. The satellite imaged the scene 20 times in a row. Each image is 200 pixels × 150 pixels with a pixel interval of 1m. The vertical baseline between satellites is randomly distributed between -500m and 500m, as shown in the figure. Figure 6 shown.

[0121] Table 1

[0122]

[0123]

[0124] Figure 7 The paper demonstrates the use of this method for data collection, and the use of compressed sensing combined with generalized likelihood ratio detection to perform three-dimensional imaging of the image, and then detects the number of overlaps within each pixel. Figure 8 The three-dimensional imaging results obtained using the traditional three-dimensional positioning model (i.e. calculating the elevation direction based on the incident angle) are shown. It can be seen that the three-dimensional imaging results of the building have a large deviation from the setting situation, and the side of the building is obviously tilted. Fig. 9 The three-dimensional imaging results obtained using this method are shown. It can be seen that the three-dimensional imaging results of the building are consistent with the setting conditions (there is no obvious tilt on the side of the building), and the three-dimensional imaging results are very close to the simulation setting scene, thereby proving the effectiveness of this method.

[0125] The three-dimensional imaging method based on the squint mode SAR tomography data of the present invention obtains N accurate ephemeris of repeated orbit satellites, obtains the scene center position and the repeated orbit satellite position at the repeated orbit aperture center moment of each repeated orbit satellite trajectory according to the accurate ephemeris of the repeated orbit satellite, N is a positive integer; selects any one of the repeated orbit satellite trajectories as a reference trajectory k0; calculates the repeated orbit satellite trajectory k0 according to the scene center position and the repeated orbit satellite position at the repeated orbit aperture center moment of the reference trajectory k0 iThe geometric correlation coefficient of the image pair with the reference trajectory k0, where i=1, 2, ..., N-1; according to the re-orbit satellite trajectory k i The precise ephemeris of the satellite is used to calculate the trajectory k of the re-orbit satellite. i to the effective vertical baseline at the center of the re-orbit aperture of the reference trajectory k0; according to the re-orbit satellite trajectory k i The geometric correlation coefficient of the image pair with the reference trajectory k0 and the effective vertical baseline are used to calculate the re-orbit satellite trajectory k i CRLB of the height inversion accuracy; Select the heavy orbit satellite trajectory k i The optimal position of the CRLB with the height inversion accuracy is taken as the re-orbit satellite trajectory k i The center of the re-orbit aperture is used to collect the re-orbit satellite trajectory k i Repeat the above steps until all the heavy orbit satellite tracks k i The center of the heavy track aperture is optimized; the target elevation is estimated using the compressed sensing method, and the three-dimensional position of the target in the geodetic coordinate system is obtained according to the elevation direction of the squint mode. The geometric correlation coefficients of each position on the trajectory and the main track are accurately predicted through SAR observation geometry, and the orbit position with the best theoretical elevation inversion accuracy is obtained, the optimal tomographic data acquisition is completed, and the accuracy of tomographic three-dimensional imaging is improved.

[0126] Although the embodiments disclosed in the present invention are as above, the above contents are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A three-dimensional imaging method based on squint mode SAR tomography data, characterized in that: The method comprises: Obtain N re-orbit satellite precise ephemerides, and obtain the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of each re-orbit satellite trajectory according to the re-orbit satellite precise ephemerides, where N is a positive integer; Select any one of the heavy orbit satellite trajectories as the reference trajectory k0; The re-orbit satellite trajectory k is calculated based on the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of the reference trajectory k0. i The geometric correlation coefficient of the image pair with the reference trajectory k0, where i=1, 2, ..., N-1; According to the re-orbit satellite trajectory k i The precise ephemeris of the satellite is used to calculate the trajectory k of the re-orbit satellite. i The effective vertical baseline to the center of the re-track aperture of the reference trajectory k0; According to the re-orbit satellite trajectory k i The geometric correlation coefficient of the image pair with the reference trajectory k0 and the effective vertical baseline are used to calculate the re-orbit satellite trajectory k i CRLB of elevation inversion accuracy; Select the heavy orbit satellite trajectory k i The optimal position of the CRLB with the height inversion accuracy is taken as the re-orbit satellite trajectory k i The center of the re-orbit aperture is used to collect the re-orbit satellite trajectory k i SAR images; Repeat the above steps until all heavy orbit satellite tracks k i The heavy rail aperture center optimization is completed; The compressed sensing method is used to estimate the target elevation, and the three-dimensional position of the target in the geodetic coordinate system is obtained according to the elevation direction of the squint mode.

2. The three-dimensional imaging method according to claim 1, characterized in that: The re-orbit satellite trajectory k is calculated based on the scene center position and the re-orbit satellite position at the re-orbit aperture center moment of the reference trajectory k0. i The geometric correlation coefficient of the image pair with the reference trajectory k0 includes: In the re-orbit satellite trajectory k i On the search, the scene center position at the moment of the heavy track aperture center of the reference trajectory k0 is found Recent Location At the stated location As the center, select the satellite trajectory intervals of 2L before and after, and evenly divide the satellite trajectory interval into M points; For the mth point, according to the re-orbit satellite trajectory k i The precise ephemeris is used to obtain the velocity of the point m The unit vector of the line of sight direction is calculated according to the position of the re-orbit satellite at the center of the re-orbit aperture of the reference trajectory k0 Calculate the unit vector The projection matrix for the normal vector in, is the re-orbit satellite position at the re-orbit aperture center of the reference trajectory k0; According to the projection matrix P k0 Calculate the ground normal vector And calculate the ground normal vector The projection matrix for the normal vector According to the unit vector Projection Matrix P k0 , ground normal vector The projection matrix Pg calculates the trajectory k of the re-orbit satellite i The geometric correlation coefficient γ of the image pair with the reference trajectory k0.

3. The three-dimensional imaging method according to claim 1, characterized in that: The geometric correlation coefficient Where c is the speed of light, λ is the wavelength of the radar payload, and B is the bandwidth of the transmitted signal. represents the slant distance from the center of the heavy track aperture of the reference trajectory k0 to the center of the scene; Represents the reference trajectory k0 and the current re-orbit satellite trajectory k i The difference between the sight unit vector from the current position to the center of the scene, express The projection on the ground, represents the projection of the synthetic aperture vector of the reference trajectory k0 on the ground; They respectively represent the ground equidistant direction and the equal Doppler direction of the reference trajectory k0 imaging.

4. The three-dimensional imaging method according to claim 3, characterized in that: The re-orbit satellite trajectory k i The precise ephemeris of the satellite is used to calculate the trajectory k of the re-orbit satellite. i The effective vertical baseline to the center of the re-track aperture of the reference trajectory k0 includes: Calculate the current re-orbit satellite trajectory k i The position of the re-orbit satellite at the time of the re-orbit aperture center The re-orbit satellite position at the time of the re-orbit aperture center of the reference trajectory k0 Baseline The effective vertical baseline 5. The three-dimensional imaging method according to claim 4, characterized in that: The elevation inversion accuracy Where γ is the trajectory k of the re-orbit satellite i The geometric correlation coefficient of the image pair with the reference trajectory k0, λ is the radar payload wavelength, b n is the trajectory k of the heavy orbit satellite i The effective vertical baseline at the time instant of the re-track aperture center to the reference trajectory k0.

Images