A method and device for three-dimensional positioning of a multi-angle space-borne SAR target

Abstract

The application discloses a kind of multi-angle spaceborne SAR target three-dimensional positioning method and device.Method includes: obtaining the multi-azimuth angle spaceborne SAR image sequence of simulation scene;Certain angle image in image sequence is regarded as main image, and the height estimation sensitivity of each other angle image is calculated based on main image;According to the target height range of simulation scene, determine multiple sampling heights;At each sampling height, the correlation coefficient between each target point in main image and its corresponding matching point in each other angle image is calculated;Based on height estimation sensitivity and the correlation coefficient under each sampling height, construct the single-angle height estimation probability density function of each other angle image of each target point in main image based on each single-angle height estimation probability density function, the height estimation value of each target point in main image is calculated;Based on the focusing position and height estimation value of each target point in main image, the three-dimensional position estimation result of the target point is calculated.The application can accurately invert the three-dimensional position of target.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of target positioning, in particular to a three-dimensional positioning method and device for multi-angle spaceborne SAR targets. BACKGROUND

[0002] SAR images are essentially the projection of three-dimensional electromagnetic scattering characteristics of targets on a two-dimensional imaging plane, and their core value lies in the realization of continuous perception of strategic targets in the monitoring area through all-weather and all-time remote sensing observation capabilities. However, due to the side-looking imaging characteristics of SAR systems, complex three-dimensional targets have problems such as top-bottom inversion, occlusion and overlap in SAR images, which leads to confusion of target spatial positions and distortion of geometric structures, and seriously restricts the interpretation and accurate positioning of targets.

[0003] Multi-azimuth angle observation spaceborne SAR systems form a curved trajectory in a long staring process, which is equivalent to forming discrete sampling in the elevation direction, and has the ability of three-dimensional reconstruction of target scenes. However, the existing three-dimensional multi-angle target three-dimensional positioning method uses approximate processing in the positioning process, resulting in an inaccurate elevation estimation model and low positioning accuracy. SUMMARY

[0004] The present application provides a three-dimensional positioning method and device for multi-angle spaceborne SAR targets, which can accurately invert the three-dimensional position of the target through a multi-angle SAR image sequence. The technical solution is as follows:

[0005] In one aspect, a three-dimensional positioning method for multi-angle spaceborne SAR targets is provided, the method comprising:

[0006] Obtaining a multi-azimuth angle spaceborne SAR image sequence of a simulated scene; wherein each SAR image corresponds to an azimuth angle;

[0007] Taking a certain angle image in the image sequence as a main image, and calculating the elevation estimation sensitivity of each other angle image based on the main image;

[0008] Determining a plurality of sampling heights according to the target height range of the simulated scene;

[0009] Calculating the correlation coefficient between each target point in the main image and its corresponding matching point in each other angle image at each sampling height;

[0010] Based on the elevation estimation sensitivity and the correlation coefficient at each sampling height, constructing a single-angle elevation estimation probability density function of each target point in the main image and each other angle image;

[0011] Based on each single-angle elevation estimation probability density function, calculating the height estimation value of each target point in the main image;

[0012] based on the focusing position and the height estimation value of each target point in the main image, calculate a three-dimensional position estimation result of the target point.

[0013] In another aspect, a device for three-dimensional positioning of a multi-angle spaceborne SAR target is provided, the device comprising:

[0014] an acquisition unit configured to acquire a sequence of multi-azimuth-angle spaceborne SAR images of a simulation scene, wherein each SAR image corresponds to an azimuth angle;

[0015] a first calculation unit configured to take a certain angle image in the image sequence as a main image, and calculate height estimation sensitivity of each other angle image based on the main image;

[0016] a sampling height determination unit configured to determine a plurality of sampling heights according to a target height range of the simulation scene;

[0017] a correlation coefficient calculation unit configured to calculate a correlation coefficient between each target point in the main image and a corresponding matching point of the target point in each other angle image at each sampling height;

[0018] a construction unit configured to construct a single-angle height estimation probability density function of each target point in the main image and each other angle image based on the height estimation sensitivity and the correlation coefficient at each sampling height;

[0019] a second calculation unit configured to calculate a height estimation value of each target point in the main image based on each single-angle height estimation probability density function;

[0020] a third calculation unit configured to calculate a three-dimensional position estimation result of each target point in the main image based on a focusing position and the height estimation value of the target point.

[0021] In another aspect, a computer device is provided, the computer device comprising a memory and a processor, the memory being configured to store a computer program, and the processor being configured to execute the computer program stored in the memory to implement the steps of the three-dimensional positioning method of the multi-angle spaceborne SAR target.

[0022] In another aspect, a computer readable storage medium is provided, the storage medium storing a computer program, the computer program being executed by a processor to implement the steps of the three-dimensional positioning method of the multi-angle spaceborne SAR target.

[0023] In another aspect, a computer program product is provided, comprising a computer program, the computer program being executed by a processor to implement the steps of the three-dimensional positioning method of the multi-angle spaceborne SAR target.

[0024] The embodiment of the present application provides a three-dimensional positioning method of a multi-angle spaceborne SAR target. Through a multi-angle SAR image sequence, first, the height estimation sensitivity between each angle image and a main image is calculated. Then, by using multiple sampling heights, the correlation coefficient between each target point and the matching point in the other angle image is calculated, and a corresponding height estimation probability density function is constructed. Then, by using each single-angle height estimation probability density function, the height estimation value of each target point in the main image can be accurately calculated. Finally, according to the height estimation value, the three-dimensional position of the target point is inverted. The method provided by the present application does not exist in the positioning process, has a more optimal height estimation model, and can realize higher three-dimensional positioning accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0026] Figure 1 It is a flow chart of a three-dimensional positioning method of a multi-angle spaceborne SAR target provided by an embodiment of the present application;

[0027] Figure 2 It is a structure diagram of a three-dimensional positioning device of a multi-angle spaceborne SAR target provided by an embodiment of the present application;

[0028] Figure 3 It is a hardware architecture diagram of a computer device provided by an embodiment of the present application;

[0029] Figure 4 It is a simulation scene schematic diagram provided by an embodiment of the present application;

[0030] Figure 5 It is a reconstruction result of a multi-angle three-dimensional positioning method based on a point projection geometry provided by an embodiment of the present application;

[0031] Figure 6 It is a reconstruction result of a multi-angle three-dimensional positioning method based on the method provided by the present application;

[0032] Figure 7 It is a comparison result schematic diagram of a multi-angle three-dimensional positioning method based on a point projection geometry and the target height estimation error of the method provided by the present application. DETAILED DESCRIPTION

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0034] The following describes the specific implementation of the above concept.

[0035] Please refer to Figure 1 This invention provides a three-dimensional localization method for multi-angle spaceborne SAR targets, the method comprising:

[0036] Step 100: Obtain a multi-azimuth spaceborne SAR image sequence of the simulation scene; where each SAR image corresponds to one azimuth angle.

[0037] Step 102: Take an image from a certain angle in the image sequence as the main image, and calculate the elevation estimation sensitivity of each other angle image based on the main image;

[0038] Step 104: Determine multiple sampling heights based on the target height range of the simulation scene;

[0039] Step 106: Calculate the correlation coefficient between each target point in the main image and its corresponding matching point in each other angle image at each sampling height;

[0040] Step 108: Based on the elevation estimation sensitivity and the correlation coefficient at each sampling height, construct the single-angle elevation estimation probability density function between each target point in the main image and each other angle image;

[0041] Step 110: Calculate the height estimate of each target point in the main image based on the probability density function of each single-angle elevation estimation.

[0042] Step 112: Calculate the three-dimensional position estimation result of the target point based on the focus position and height estimation value of each target point in the main image.

[0043] In this embodiment of the invention, using a multi-angle SAR image sequence, the elevation estimation sensitivity between each angle image and the main image is first calculated. Then, using multiple sampling heights, the correlation coefficient between each target point and matching points in other angle images is calculated, and a corresponding elevation estimation probability density function is constructed. Next, using each single-angle elevation estimation probability density function, the elevation estimate of each target point in the main image can be accurately calculated. Finally, the three-dimensional position of the target point is retrieved based on the elevation estimate. This method does not involve approximation during the positioning process, possesses a superior elevation estimation model, and can achieve higher three-dimensional positioning accuracy.

[0044] The following description Figure 1 The execution method for each step is shown.

[0045] First, regarding step 100:

[0046] Each SAR image's parameters include: satellite position and velocity vectors at the observation center time corresponding to the image, azimuth resolution, ground range resolution, azimuth sampling interval, ground range sampling interval, number of azimuth points, and number of range points. Furthermore, the simulation parameters for the simulation scene include the image tile length L and the scene height range [h]. min ,h max ], h min and h max These are the minimum height and the maximum height, respectively.

[0047] In addition, the number and angle range of SAR image sequences are determined according to user needs, and this application does not impose specific limitations.

[0048] Then, regarding step 102:

[0049] Any image at any angle can be used as the main image. This application prefers to use the image corresponding to the first azimuth angle as the main image (i.e., the first image in the image sequence) to ensure higher calculation accuracy.

[0050] In some implementations, step 102 includes:

[0051] Step B1: Set the target height unit vector and intermediate matrix;

[0052] Step B2: Based on the satellite position vector and velocity vector, target altitude unit vector and intermediate matrix at the observation center time corresponding to each angle image, calculate the solution of the forward positioning equation corresponding to each angle image;

[0053] Step B3: Based on the solution of the forward positioning equation corresponding to each angle image, calculate the focal position of the target height unit vector in each angle image;

[0054] Step B4: Calculate the azimuth offset and ground distance offset between the focus position of the target height unit vector in the main image and its focus position in other angle images;

[0055] Step B5: Calculate the elevation estimation sensitivity for each other angle image based on azimuth offset and distance offset.

[0056] In step B1, the target height unit vector Q is preferably:

[0057] The preferred intermediate matrix A is:

[0058] Of course, users may use other matrices, and this application is not limited to them.

[0059] In step B2, the solution to the forward positioning equation corresponding to each angle image is calculated using the following formula:

[0060] α k =[V k (2); -V k (1); 0]

[0061]

[0062] In the formula, α k β k and c k These represent the first solution, the second solution, and the constant coefficient of the forward positioning equation corresponding to the angle k image; V k and S k Let be the satellite's velocity vector and position vector at the observation center time corresponding to the angle k image, respectively. S k (1) S k (2) and S k (3) These are the position components in the x, y, and z directions, respectively; V k (1) V k (2) and V k (3) The velocity components in the x, y, and z directions are respectively; k = 1, 2, ... N look N look This represents the number of azimuth angles, i.e., the number of image sequences.

[0063] In step B3, the focal position of the target height unit vector in each angle image is calculated using the following formula:

[0064] P k =c k ·α k +β k

[0065] In the formula, P k Let Q be the focal point of the target height unit vector in the angle k image. P k (1) P k (2) and P k (3) The components in the x, y, and z directions are respectively.

[0066] In step B4, the azimuth offset and the distance offset are calculated using the following formulas:

[0067] Δx 1-k =P k (1)-P1(1),Δy 1-k =P k (2)-P1(2)

[0068] In the formula, Δx 1-k and Δy 1-k These represent the azimuth offset and ground distance offset between the focal position of the target height unit vector Q in the angle 1 image (main image) and the focal position in the angle k image, respectively.

[0069] In step B5, the elevation estimation sensitivity for each other angle image is calculated using the following formula:

[0070]

[0071] In the formula, ε k Sensitivity for elevation estimation of the angle k image; ρ kx and ρ ky These represent the azimuth resolution and ground distance resolution of the image at angle k, where k = 2, 3, ..., N. look .

[0072] Then, for step 104, the following formula is used to determine multiple sampling heights:

[0073] h(n) = h min +0.1·n,n=1,2,…,round[10·(h max -h min )]

[0074] In the formula, h(n) is the nth sampling height; round(·) is the rounding operation.

[0075] For step 106, the following are included:

[0076] For each sampling height, perform the following:

[0077] Step C1: Set the height of each target point in the main image to be equal to the sampling height, and calculate the sampling 3D position of each target point at the sampling height;

[0078] Step C2: Calculate the focal position of each target point in each other angle image when it is in its sampled 3D position based on the forward localization equation;

[0079] Step C3: Extract a first image slice of a preset size around each target point in the main image, and a second image slice of a preset size around the focal position of each target point in each other angle image;

[0080] Step C4: Calculate the correlation coefficient between the first image slice and the second image slice corresponding to each target point, and use the correlation coefficient as the correlation coefficient between each target point and its corresponding matching point in each other angle image at the sampling height.

[0081] In some implementations, step C1 includes:

[0082] For each target point in the main image, perform the following:

[0083] Step D1: Calculate the focal position of the target point in the main image based on the pixel position of the target point in the main image, the number of azimuth points, the number of range points, the azimuth sampling interval, and the ground distance sampling interval.

[0084] Step D2: Based on the focus position, the sampling height, and the satellite position vector and velocity vector at the observation center time corresponding to the main image, calculate the solution of the reverse positioning equation corresponding to the target point;

[0085] Step D3: Based on the solution of the reverse positioning equation, calculate the sampling three-dimensional position of the target point at the sampling height.

[0086] In step D1, taking the i-th target point as an example, the focal position of the target point in the main image is calculated using the following formula:

[0087]

[0088] In the formula, P 1i Let i be the focal position of the target point i in the main image; and δ represents the pixel position of target point i in the main image; x and δ y These represent the azimuth sampling interval and the ground distance sampling interval of the SAR image, respectively; N x and N y These represent the number of points in the azimuth direction and the number of points in the range direction of the SAR image, respectively.

[0089] In step D2, let the sampling height of target point i be h. n Then, the solution to the reverse positioning equation corresponding to target point i is calculated using the following formula:

[0090]

[0091]

[0092] In the formula, ξ i and d i These are the first solution, the second solution, and the constant coefficient of the reverse positioning equation corresponding to target point i; P 1i (1) and P 1i (2) P respectively 1i Components in the x and y directions.

[0093] In step D3, the sampling three-dimensional position of target point i at the sampling height is calculated using the following formula:

[0094]

[0095] In the formula, T i The sampling three-dimensional position of target point i at this sampling height

[0096] In step C2, the target point i at the sampling three-dimensional position T is calculated using the following formula. i At that time, the focal position of target point i in the angle k image is:

[0097] α ki =[V k (2); -V k (1); 0]

[0098]

[0099] P ki =c ki ·α ki +β ki

[0100] In the formula, P ki For target point i at sampling 3D position T i At that time, its focal position in the angle k image, and These are the components in the x and y directions, respectively; α ki β ki and c ki The target point i is located at the sampling 3D position T. i At that time, the first solution, the second solution, and the constant coefficients of the forward positioning equation corresponding to the angle k image; k = 2, 3, ..., N look .

[0101] In step C3, the preset size is 2L+1, and the target point i is the first image slice. The center of each target point i is the focal position of the angle image k, which is the second image slice. The center.

[0102] In step C4, the following formula is used to calculate the sampling height h. n Below, the correlation coefficient between the first and second image slices corresponding to target point i:

[0103]

[0104] In the formula, and , respectively, are the mean values ​​of the first and second image slices corresponding to target point i.

[0105] In some implementations, step 108 includes:

[0106] For each target point in each other angle image and in the main image, perform the following:

[0107] S1, Based on the correlation coefficient between the target point and the matching point corresponding to the current angle image at each sampling height, plot the discrete correlation coefficient curve, and then execute S2;

[0108] S2, solve for the maximum peak value in the current correlation coefficient curve and the peak height corresponding to the maximum peak value, and construct the current correlation coefficient model at the peak height based on the elevation estimation sensitivity of the current angle image, and then execute S3;

[0109] S3: Remove the maximum peak value and the discrete points within a preset range around the peak value from the current correlation coefficient model curve to obtain the updated correlation coefficient curve. Determine whether the maximum peak value still exists in the updated correlation coefficient curve. If yes, use the updated correlation coefficient curve as the current correlation coefficient curve and return to execute S2. If no, use the current correlation coefficient model as the final correlation coefficient model and execute S4.

[0110] S4. Normalize the final correlation coefficient model to obtain the single-angle elevation estimation probability density function of the target point and the current angle image.

[0111] In this step, the formula for calculating the final correlation coefficient model for images at each angle is as follows:

[0112]

[0113] The probability density functions for single-angle elevation estimation of images at various angles are as follows:

[0114]

[0115] In the formula, the subscript k represents the image corresponding to the azimuth angle k; Coh′ k (h) represents the final correlation coefficient model corresponding to the angle k image; γ km Let h be the correlation coefficient of the m-th maximum peak point, where m = 1, 2, ..., M, and M is the total number of peak points; h is the high-order sequence composed of each sampling height; h km To be with γ km Corresponding peak height; ε k Sensitivity for elevation estimation corresponding to the angle k image; {·} mainlobe This refers to the operation of retrieving the main lobe region of the sinc function; pdf k (h) is the elevation estimation probability density function corresponding to the angle k image; h min and h max These are the minimum and maximum heights in the high-order sequence, respectively.

[0116] In some implementations, step 110 includes:

[0117] Based on the probability density function of each single-angle elevation estimation, the following formula is used to construct the joint probability density function for multiple angles:

[0118]

[0119] Solve for the maximum likelihood solution of the joint probability density function of multiple angles, and calculate the height estimate of each target point in the main image;

[0120] In the formula, pdf multi (h) is the joint probability density function from multiple angles; pdf k (h) is the probability density function for elevation estimation of the angle k image; N look This represents the total number of azimuth angles.

[0121] In addition, the height estimate of target point i in the main image is calculated using the following formula.

[0122]

[0123] In some implementations, for step 112, the three-dimensional position estimation result of target point i is calculated using the following formula:

[0124] ζ' i = [-V1(2); V1(1); 0]

[0125]

[0126] T' i =d' i ·ζ i +ξ' i

[0127] In the formula, ζ' i ξ' i and d' i The estimated heights at target point i are respectively: At that time, the first solution, the second solution, and the constant coefficient of the corresponding reverse positioning equation; T' i The height estimate of target point i is The three-dimensional position estimation results at that time.

[0128] In summary, by using the above process to calculate the 3D position estimation results of each target point in the main image, a 3D point cloud map can be generated, and the 3D positioning result map of the simulation scene can be obtained.

[0129] To demonstrate the beneficial effects of this invention, the inventors compared the estimation results of the method of this application with the estimation results of an existing method (a multi-angle 3D positioning method based on image point projection geometry) using the following embodiments. The simulation scenario is as follows: Figure 4 As shown in the figure, each image contains 11 target points, denoted as P1 to P11. Furthermore, the simulation parameters are shown in Table 1.

[0130] Table 1 System Simulation Parameters

[0131]

[0132] In addition, SAR images from 10 angles were used, with the image at angle 1 as the main image. The observation parameters of the SAR images at different angles are shown in Table 2.

[0133] Table 2 SAR image observation parameters at different angles

[0134]

[0135] The calculations were performed using the parameters mentioned above, and the results obtained using existing methods are as follows: Figure 5 As shown, the calculation results using the method of this application are as follows: Figure 6 As shown, the yellow box represents the true location of the target, and the dots represent the algorithm's estimation results. A direct comparison of the 3D reconstruction results of the two algorithms reveals that the height estimation error of the image point projection geometry-based method is larger, with many points located outside the true location, and its estimated maximum height also far exceeds the true maximum height of the target. In contrast, the method proposed in this application estimates the target location most closely to the true target location, which is essentially within the yellow box.

[0136] also, Figure 7The elevation estimation errors of target points are compared between the image point projection geometry-based method and the method proposed in this application. A comparison of the elevation estimation errors of each target point shows that the elevation estimation accuracy of the method proposed in this application is superior, with the error generally controlled within 2m. In contrast, the elevation estimation error of the image point projection geometry-based method exceeds 30m, resulting in significant positioning errors.

[0137] Therefore, the calculation results of the method in this application are closer to the actual target position, and the calculation accuracy is higher.

[0138] like Figure 2 , Figure 3 As shown, this embodiment of the invention provides a three-dimensional positioning device for multi-angle spaceborne SAR targets. The device embodiment can be implemented through software, hardware, or a combination of both. From a hardware perspective, as... Figure 2 The diagram shown is a hardware architecture diagram of a computing device housing a multi-angle spaceborne SAR target three-dimensional positioning device provided in an embodiment of the present invention, except for... Figure 2 In addition to the processor, memory, network interface, and non-volatile memory shown, the computing device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing packets. Taking software implementation as an example, such as... Figure 3 As shown, a device in a logical sense is formed by the CPU of the computing device in which it is located reading the corresponding computer program from the non-volatile memory into the memory for execution.

[0139] Please refer to Figure 3 This application provides a three-dimensional positioning device for multi-angle spaceborne SAR targets, comprising:

[0140] The acquisition unit 300 is used to acquire a multi-azimuth spaceborne SAR image sequence of the simulation scene; wherein each SAR image corresponds to one azimuth angle;

[0141] The first calculation unit 302 is used to take an image at a certain angle in the image sequence as the main image and calculate the elevation estimation sensitivity of each other angle image based on the main image.

[0142] The sampling height determination unit 304 is used to determine multiple sampling heights based on the target height range of the simulation scene;

[0143] The correlation coefficient calculation unit 306 is used to calculate the correlation coefficient between each target point in the main image and its corresponding matching point in each other angle image at each sampling height;

[0144] Construction unit 308 is used to construct the single-angle elevation estimation probability density function between each target point in the main image and each other angle image based on the elevation estimation sensitivity and the correlation coefficient at each sampling height;

[0145] The second calculation unit 310 is used to calculate the height estimate of each target point in the main image based on the probability density function of each single-angle elevation estimation.

[0146] The third calculation unit 312 is used to calculate the three-dimensional position estimation result of the target point based on the focus position and height estimation value of each target point in the main image.

[0147] In some implementations, the first computing unit 302 is used to perform the following operations:

[0148] Set the target height unit vector and intermediate matrix;

[0149] Based on the satellite position vector and velocity vector, target altitude unit vector and intermediate matrix at the observation center time corresponding to each angle image, the solution of the forward positioning equation corresponding to each angle image is calculated respectively;

[0150] Based on the solution of the forward localization equation corresponding to each angle image, calculate the focal position of the target height unit vector in each angle image;

[0151] Calculate the azimuth offset and ground distance offset between the focal position of the target height unit vector in the main image and its focal position in other angle images;

[0152] Based on azimuth offset and distance offset, the elevation estimation sensitivity of each other angle image is calculated.

[0153] In some implementations, the correlation coefficient calculation unit 306 is used to perform the following operations:

[0154] For each sampling height, perform the following:

[0155] Set the height of each target point in the main image to be equal to the sampling height, and calculate the sampling 3D position of each target point at the sampling height;

[0156] The focus position of each target point in each other angle image is calculated based on the positive localization equation when it is in its sampled 3D position;

[0157] Extract a first image slice of a preset size around each target point in the main image, and a second image slice of a preset size around the focal position of each target point in each other angle image;

[0158] Calculate the correlation coefficient between the first image slice and the second image slice corresponding to each target point, and use this correlation coefficient as the correlation coefficient between each target point and its corresponding matching point in each other angle image at the sampling height.

[0159] In some implementations, the correlation coefficient calculation unit 306 performs the following operations when calculating the three-dimensional sampling position of each target point at the sampling height:

[0160] For each target point in the main image, perform the following:

[0161] Calculate the focal position of the target point in the main image based on the pixel position of the target point in the main image, the number of points in the azimuth direction, the number of points in the range direction, the azimuth sampling interval, and the ground distance sampling interval;

[0162] Based on the focus position, the sampling height, and the satellite position vector and velocity vector at the observation center time corresponding to the main image, calculate the solution to the reverse positioning equation corresponding to the target point;

[0163] Based on the solution of the reverse positioning equation, the sampling three-dimensional position of the target point at the sampling height is calculated.

[0164] In some implementations, the construction unit 308 is used to perform the following operations:

[0165] For each target point in each other angle image and in the main image, perform the following:

[0166] S1, Based on the correlation coefficient between the target point and the matching point corresponding to the current angle image at each sampling height, plot the discrete correlation coefficient curve, and then execute S2;

[0167] S2, solve for the maximum peak value in the current correlation coefficient curve and the peak height corresponding to the maximum peak value, and construct the current correlation coefficient model at the peak height based on the elevation estimation sensitivity of the current angle image, and then execute S3;

[0168] S3: Remove the maximum peak value and the discrete points within a preset range around the peak value from the current correlation coefficient model curve to obtain the updated correlation coefficient curve. Determine whether the maximum peak value still exists in the updated correlation coefficient curve. If yes, use the updated correlation coefficient curve as the current correlation coefficient curve and return to execute S2. If no, use the current correlation coefficient model as the final correlation coefficient model and execute S4.

[0169] S4. Normalize the final correlation coefficient model to obtain the single-angle elevation estimation probability density function of the target point and the current angle image.

[0170] In some implementations, the formula for calculating the final correlation coefficient model of images from different angles is as follows:

[0171]

[0172] The probability density functions for single-angle elevation estimation of images at various angles are as follows:

[0173]

[0174] In the formula, the subscript k represents the image corresponding to the azimuth angle k; Coh k ′(h) represents the final correlation coefficient model corresponding to the image at angle k; γ km Let h be the correlation coefficient of the m-th maximum peak point, where m = 1, 2, ..., M, and M is the total number of peak points; h is the high-order sequence composed of each sampling height; h km To be with γ km Corresponding peak height; ε k Sensitivity for elevation estimation corresponding to the angle k image; {·} mainlobe This refers to the operation of retrieving the main lobe region of the sinc function; pdf k (h) is the elevation estimation probability density function corresponding to the angle k image; h min and h max These are the minimum and maximum heights in the high-order sequence, respectively.

[0175] In some implementations, the second computing unit is used to perform the following operations:

[0176] Based on the probability density function of each single-angle elevation estimation, the following formula is used to construct the joint probability density function for multiple angles:

[0177]

[0178] Solve for the maximum likelihood solution of the joint probability density function of multiple angles, and calculate the height estimate of each target point in the main image;

[0179] In the formula, pdf multi (h) is the joint probability density function from multiple angles; pdf k (h) is the probability density function for elevation estimation of the angle k image; N look This represents the total number of azimuth angles.

[0180] It should be noted that the three-dimensional positioning device for multi-angle spaceborne SAR targets provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the three-dimensional positioning device for multi-angle spaceborne SAR targets provided in the above embodiments and the three-dimensional positioning method embodiments for multi-angle spaceborne SAR targets belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0181] Embodiments of this application also provide a computer device, please refer to... Figure 3The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, at least one program, code set or instruction set being loaded and executed by the processor to implement the three-dimensional positioning method for multi-angle spaceborne SAR targets provided in the above-described method embodiments.

[0182] The embodiments of this application also provide a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the three-dimensional positioning method for multi-angle spaceborne SAR targets provided in the above-described method embodiments.

[0183] Embodiments of this application also provide a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium and executes the computer program, causing the computer device to perform the three-dimensional positioning method for multi-angle spaceborne SAR targets as described in any of the above embodiments.

[0184] For ease of description, the above systems or devices are described separately as various modules or units based on their functions. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware components.

[0185] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0186] Finally, it should be noted that in this document, relational terms such as first, second, third, and fourth are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0187] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A three-dimensional positioning method for multi-angle spaceborne SAR targets, characterized in that, The method includes: Acquire a multi-azimuth spaceborne SAR image sequence of the simulation scene; where each SAR image corresponds to one azimuth angle; The image at a certain angle in the image sequence is taken as the master image, and the elevation estimation sensitivity of each other angle image is calculated based on the master image; Based on the target height range of the simulation scene, multiple sampling heights are determined; Calculate the correlation coefficient between each target point in the main image and its corresponding matching point in each other angle image at each sampling height; Based on the elevation estimation sensitivity and the correlation coefficient at each sampling height, a single-angle elevation estimation probability density function is constructed for each target point in the main image and each other angle image. Based on the probability density function of each single-angle elevation estimation, the height estimate of each target point in the main image is calculated; Based on the focus position and height estimate of each target point in the main image, calculate the three-dimensional position estimate of the target point; The construction of a single-angle elevation estimation probability density function for each target point in the main image and each other angle image, based on the elevation estimation sensitivity and the correlation coefficient at each sampling height, includes: For each other angle image and each target point in the main image, perform the following: S1, Based on the correlation coefficient between the target point and the matching point corresponding to the current angle image at each sampling height, plot the discrete correlation coefficient curve, and then execute S2; S2, solve for the maximum peak value in the current correlation coefficient curve and the peak height corresponding to the maximum peak value, and construct the current correlation coefficient model at the peak height based on the elevation estimation sensitivity of the current angle image, and then execute S3; S3: Remove the maximum peak value and the discrete points within a preset range around the peak value from the current correlation coefficient model curve to obtain the updated correlation coefficient curve. Determine whether the maximum peak value still exists in the updated correlation coefficient curve. If yes, use the updated correlation coefficient curve as the current correlation coefficient curve and return to execute S2. If no, use the current correlation coefficient model as the final correlation coefficient model and execute S4. S4, normalize the final correlation coefficient model to obtain the single-angle elevation estimation probability density function of the target point and the current angle image; The formula for calculating the final correlation coefficient model for images from different angles is as follows: The probability density functions for single-angle elevation estimation of images at various angles are as follows: In the formula, the subscript k Indicates azimuth angle k The corresponding image; For angle k The final correlation coefficient model corresponding to the image; For the first m The correlation coefficient of the maximum peak point m =1,2…… M , M This represents the total number of peak points; h It is a high-order sequence composed of each sampling height; To and Corresponding peak height; Sensitivity for elevation estimation corresponding to the angle k image; {·} mainlobe This is an operation to retrieve the main lobe region of the sinc function; For angle k The elevation estimation probability density function corresponding to the image; and These are the minimum and maximum heights in the high-order column, respectively; The step of calculating the height estimate of each target point in the main image based on the probability density function of each single-angle elevation estimation includes: Based on the probability density function of each single-angle elevation estimation, the following formula is used to construct the joint probability density function for multiple angles: Solve for the maximum likelihood solution of the multi-angle joint probability density function, and calculate the height estimate of each target point in the main image; In the formula, It is a joint probability density function from multiple perspectives; For angle k The probability density function for elevation estimation of an image; This represents the total number of azimuth angles.

2. The method according to claim 1, characterized in that, The process of calculating the elevation estimation sensitivity for each other angle image based on the main image includes: Set the target height unit vector and intermediate matrix; Based on the satellite position vector and velocity vector at the observation center time corresponding to each angle image, the target height unit vector, and the intermediate matrix, the solution of the forward positioning equation corresponding to each angle image is calculated respectively; Based on the solution of the forward positioning equation corresponding to each angle image, the focal position of the target height unit vector in each angle image is calculated respectively; Calculate the azimuth offset and ground distance offset between the focal position of the target height unit vector in the main image and its focal position in other angle images; Based on the azimuth offset and the ground distance offset, the elevation estimation sensitivity of each other angle image is calculated.

3. The method according to claim 1, characterized in that, The calculation of the correlation coefficient between each target point in the main image and its corresponding matching point in each other angle image at each sampling height includes: For each sampling height, perform the following: Set the height of each target point in the main image to be equal to the sampling height, and calculate the sampling three-dimensional position of each target point at the sampling height; The focus position of each target point in each other angle image is calculated based on the positive localization equation when it is in its sampled 3D position; Extract a first image slice of a preset size around each target point in the main image, and a second image slice of the preset size around the focal position of each target point in each other angle image; Calculate the correlation coefficient between the first image slice and the second image slice corresponding to each target point, and use this correlation coefficient as the correlation coefficient between each target point and its corresponding matching point in each other angle image at the sampling height.

4. The method according to claim 3, characterized in that, The calculation of the sampling three-dimensional position of each target point at the sampling height includes: For each target point in the main image, the following is performed: Calculate the focal position of the target point in the main image based on the pixel position of the target point in the main image, the number of azimuth points, the number of range points, the azimuth sampling interval, and the ground distance sampling interval; Based on the focal position, the sampling height, and the satellite position vector and velocity vector at the observation center time corresponding to the main image, calculate the solution of the reverse positioning equation corresponding to the target point; Based on the solution of the reverse positioning equation, the sampling three-dimensional position of the target point at the sampling height is calculated.

5. A three-dimensional positioning device for multi-angle spaceborne SAR targets, characterized in that, The apparatus for implementing the method according to any one of claims 1-4 comprises: The acquisition unit is used to acquire a sequence of multi-azimuth spaceborne SAR images of the simulation scene; where each SAR image corresponds to one azimuth angle. The first calculation unit is used to take an image at a certain angle in the image sequence as the main image, and calculate the elevation estimation sensitivity of each other angle image based on the main image; The sampling height determination unit is used to determine multiple sampling heights based on the target height range of the simulation scene; The correlation coefficient calculation unit is used to calculate the correlation coefficient between each target point in the main image and its corresponding matching point in each other angle image at each sampling height; The construction unit is used to construct a single-angle elevation estimation probability density function between each target point in the main image and each other angle image, based on the elevation estimation sensitivity and the correlation coefficient at each sampling height. The second calculation unit is used to calculate the height estimate of each target point in the main image based on the probability density function of each single-angle elevation estimation. The third calculation unit is used to calculate the three-dimensional position estimation result of each target point based on the focus position and height estimation value of each target point in the main image.

6. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to implement the steps of the method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method described in any one of claims 1-4.

Images