Dual-frequency tomographic synthetic aperture radar three-dimensional imaging method, system, equipment and medium

Abstract

The present invention discloses a dual-frequency tomographic synthetic aperture radar (SAR) three-dimensional imaging method, system, device, and medium. The method uses a dual-frequency tomographic synthetic aperture radar (SAR) system to acquire low-frequency and high-frequency image data of a target area. The low-frequency image data undergoes neighborhood pixel-by-pixel joint processing to obtain a low-frequency data stack, and the high-frequency image data undergoes neighborhood pixel-by-pixel joint processing to obtain a high-frequency data stack. Based on the principle of dual-frequency tomography, the low-frequency and high-frequency data stacks are fused to obtain a dual-frequency data stack. A distributed compressed sensing algorithm is used to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector. A three-dimensional point cloud is generated based on the elevation scattering coefficient vector to obtain a three-dimensional image of the target area. The method provided in this application can reduce the number of required passes, effectively highlight the location of scatterers, and significantly improve the accuracy of resolving elevation ambiguities.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional imaging technology, and in particular to a dual-frequency tomographic synthetic aperture radar three-dimensional imaging method, system, equipment and medium. Background Art

[0002] The tomosynthetic aperture radar (TAR) system forms a height-dimensional synthetic aperture by observing multiple upward trajectories. This allows it to distinguish multiple scatterers at different altitudes within the same pixel, enabling it to create three-dimensional images of the observed target. Currently, the data used primarily comes from single-satellite, single-frequency radar re-orbits, which typically require dozens of repeated flights. Single-frequency TSAR reconstructions face challenges such as insufficient observational information, elevation ambiguity, and low three-dimensional imaging accuracy. Satellites equipped with dual-frequency radars that can operate simultaneously can present different details and feature information, but they still suffer from the high number of flybys, poor deambiguation capabilities at low signal-to-noise ratios, and low three-dimensional imaging accuracy.

[0003] Therefore, how to solve the problems of many flybys and low three-dimensional imaging accuracy in the existing tomographic synthetic aperture radar three-dimensional imaging method has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0004] The present invention provides a dual-frequency tomographic synthetic aperture radar three-dimensional imaging method, system, device and medium to solve the technical problems of the existing tomographic synthetic aperture radar three-dimensional imaging method, such as the large number of flybys and low three-dimensional imaging accuracy, thereby achieving the effect of improving the three-dimensional imaging accuracy while reducing the number of flybys.

[0005] In a first aspect, the present invention provides a tomographic synthetic aperture radar three-dimensional imaging method, the method comprising: acquiring low-frequency image data and high-frequency image data of a target area using a dual-frequency tomographic synthetic aperture radar system, performing neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and performing neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack;

[0006] Based on the principle of dual-frequency tomography, the low-frequency data stack and the high-frequency data stack are fused to obtain a dual-frequency data stack;

[0007] A distributed compressed sensing algorithm is used to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector;

[0008] A three-dimensional point cloud is generated according to the elevation scattering coefficient vector to obtain a three-dimensional image of the target area.

[0009] Preferably, performing neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and performing neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack, comprises:

[0010] In the low-frequency image data, selecting a preset number of first neighboring pixels within a preset range for each range-azimuth pixel to construct a low-frequency data stack;

[0011] In the high-frequency image data, selecting a preset number of second neighboring pixels within the preset range for each range-azimuth pixel to construct a high-frequency data stack;

[0012] The first neighborhood pixels and the second neighborhood pixels include vertically adjacent pixels, horizontally adjacent pixels, and diagonally adjacent pixels.

[0013] Preferably, the low-frequency data stack and the high-frequency data stack are fused based on the dual-frequency tomography principle to obtain a dual-frequency data stack, including:

[0014] Based on the equivalent synthetic aperture principle, a single-pixel continuous signal model is constructed, and the single-pixel continuous signal model is expanded into a dual-pixel continuous signal model;

[0015] Discretizing the dual-pixel continuous signal model to obtain a dual-frequency observation signal matrix expression;

[0016] Performing baseline equivalent processing on the dual-frequency observation signal matrix expression;

[0017] The dual-frequency observation signal matrix expression after baseline equivalent processing is used to fuse the low-frequency data stack and the high-frequency data stack to obtain a dual-frequency data stack.

[0018] Preferably, the dual-frequency data stack is represented as:

[0019]

[0020] Where g1(λ) represents the observation vector of the first pixel in the low-frequency data stack or the high-frequency data stack, g2(λ) represents the observation vector of the second pixel in the low-frequency data stack or the high-frequency data stack, and g W (λ) represents the observation vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack, W represents the number of pixels in the low-frequency data stack or the high-frequency data stack, R1(λ) represents the observation vector of the first pixel in the low-frequency data stack or the high-frequency data stack, R2(λ) represents the observation vector of the second pixel in the low-frequency data stack or the high-frequency data stack, and R W(λ) represents the observation vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack, ⊙ represents the Hadamard product of the matrix, γ1(λ) represents the elevation scattering coefficient of the first pixel in the low-frequency data stack or the high-frequency data stack, γ2(λ) represents the elevation scattering coefficient of the second pixel in the low-frequency data stack or the high-frequency data stack, and γ W (λ) represents the vertical scattering coefficient of the Wth pixel in the low-frequency data stack or the high-frequency data stack, λ represents the radar wavelength, λ1 represents the radar wavelength for collecting low-frequency image data, λ2 represents the radar wavelength for collecting high-frequency image data, ε1(λ) represents the noise vector of the first pixel in the low-frequency data stack or the high-frequency data stack, ε2(λ) represents the noise vector of the second pixel in the low-frequency data stack or the high-frequency data stack, ε W (λ) represents the noise vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack.

[0021] Preferably, the method of using a distributed compressed sensing algorithm to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector includes:

[0022] Constructing a dual-frequency joint solution model, initializing the number of iterations of the dual-frequency joint solution model, the orthogonalized vertical scattering coefficient, the basis vector index set, the dictionary vector set and the observation vector residual;

[0023] In an iterative update process, the dual-frequency data stack is input into the initialized dual-frequency joint solution model, a target basis vector that maximizes the sum of the projections of the observation vector residuals is calculated, and an index of the target basis vector is added to the basis vector index set;

[0024] Selecting a basis vector through the basis vector index set, selecting a dictionary vector from the dictionary vector set, orthogonalizing the selected basis vector and the dictionary vector to obtain an orthogonalized basis vector, and incorporating the orthogonalized basis vector into the dictionary vector set;

[0025] Calculate and update the orthogonalized vertical scatter coefficient and the observation vector residual according to the orthogonalized basis vector;

[0026] determining whether all pixels of the dual-frequency data stack meet the convergence requirement based on the observation vector and the updated observation vector residual, and if not, performing the iterative update process until the convergence requirement is met;

[0027] Orthogonal triangular decomposition is used to de-orthogonalize the dictionary vector set and the orthogonalized elevation scattering coefficient after the convergence requirement is met, so as to obtain an elevation scattering coefficient vector.

[0028] Preferably, the orthogonal triangular decomposition is used to deorthogonalize the dictionary vector set and the orthogonalized elevation scattering coefficient after the convergence requirement is met to obtain the elevation scattering coefficient vector, including:

[0029] Uniquely decomposing the dictionary vector after meeting the convergence requirement into an orthogonal matrix and an upper triangular matrix;

[0030] Based on the orthogonal matrix and the upper triangular matrix, de-orthogonalization processing is performed on the dictionary vector set and the orthogonalized elevation scattering coefficient that meet the convergence requirement to obtain an elevation scattering coefficient vector.

[0031] Preferably, the calculation expression of the elevation scattering coefficient vector is:

[0032]

[0033] in, represents the inverse of the upper triangular matrix, Q H represents the transpose of an orthogonal matrix, represents the dictionary vector after meeting the convergence requirements, It represents the orthogonalized vertical scattering coefficient after meeting the convergence requirements.

[0034] In a second aspect, the present invention further provides a dual-frequency tomographic synthetic aperture radar 3D imaging system for implementing the above-mentioned dual-frequency tomographic synthetic aperture radar 3D imaging method, the system comprising: a data acquisition and processing unit, a data fusion unit, a dual-frequency joint solution unit, and a 3D imaging unit;

[0035] The data acquisition and processing unit is configured to acquire low-frequency image data and high-frequency image data of the target area through a dual-frequency tomography synthetic aperture radar system, perform neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and perform neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack;

[0036] The data fusion unit is configured to fuse the low-frequency data stack and the high-frequency data stack based on the dual-frequency tomography principle to obtain a dual-frequency data stack;

[0037] The dual-frequency joint solving unit is used to perform a dual-frequency joint solution on the dual-frequency data stack using a distributed compressed sensing algorithm to obtain an elevation scattering coefficient vector;

[0038] The three-dimensional imaging unit is used to generate a three-dimensional point cloud according to the elevation scattering coefficient vector to obtain a three-dimensional image of the target area.

[0039] In a third aspect, the present invention also provides a computer device, comprising a memory, a processor, and a transceiver, which are connected via a bus; the memory is used to store a set of computer program instructions and data, and transmit the stored data to the processor, and the processor executes the program instructions stored in the memory to execute the above-mentioned dual-frequency tomography synthetic aperture radar three-dimensional imaging method.

[0040] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium. When the computer program is executed, the above-mentioned dual-frequency tomography synthetic aperture radar three-dimensional imaging method is implemented.

[0041] The present invention provides a dual-frequency tomographic synthetic aperture radar three-dimensional imaging method, system, device, and medium. Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0042] (1) The neighborhood information of single-frequency data is combined to improve the signal-to-noise ratio and reduce the number of required flybys.

[0043] (2) The information contained in the combined dual-frequency data can effectively highlight the position of the scatterer, increase the unambiguous height, and further reduce the number of flights, thus greatly improving the accuracy of resolving the ambiguity in the elevation direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic diagram of the steps of a dual-frequency tomographic synthetic aperture radar three-dimensional imaging method provided by a preferred embodiment of the present invention;

[0045] Figure 2 1 is a schematic diagram of imaging geometry of a dual-frequency tomographic synthetic aperture radar system provided by a preferred embodiment of the present invention;

[0046] Figure 3 This is image data acquired by a simulated dual-frequency tomographic synthetic aperture radar system provided by a preferred embodiment of the present invention;

[0047] Figure 4 is an elevation image of a single scatterer model provided by a preferred embodiment of the present invention;

[0048] Figure 5 A preferred embodiment of the present invention provides an elevation scattering profile using the OMP algorithm (orthogonal matching pursuit algorithm) and the DCS algorithm (distributed compressed sensing algorithm) to image the X-band and C-band separately;

[0049] Figure 6It is an elevation scattering profile imaged using the Dual-OMP algorithm (dual-frequency orthogonal matching pursuit algorithm) and the Dual-DCS (dual-frequency tomographic synthetic aperture radar three-dimensional imaging method disclosed in this application) provided by a preferred embodiment of the present invention;

[0050] Figure 7 This is a schematic diagram of the experimental results of a single scatterer model provided by a preferred embodiment of the present invention. Figure 1 ;

[0051] Figure 8 This is a schematic diagram of the experimental results of a single scatterer model provided by a preferred embodiment of the present invention. Figure 2 ;

[0052] Figure 9 This is a schematic structural diagram of a dual-frequency tomographic synthetic aperture radar three-dimensional imaging system provided by a preferred embodiment of the present invention;

[0053] Figure 10 It is a structural diagram of a computer device provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following is a detailed explanation of the embodiments of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and cannot be understood as limitations on the present invention. The accompanying drawings are for reference and illustration purposes only and do not constitute a limitation on the scope of protection of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In the description of the present invention, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, the meaning of "multiple" is two or more.

[0055] In the description of the present invention, it should be noted that, unless otherwise expressly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be a communication between the two components. The terms "vertical", "horizontal", "left", "right", "up", "down" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0056] In describing the present invention, it should be noted that, unless otherwise defined, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. Those skilled in the art will understand the specific meanings of the above terms in the present invention in specific circumstances.

[0057] In an embodiment of the present invention, a tomographic synthetic aperture radar three-dimensional imaging method is provided. Figure 1 , the method comprising:

[0058] S1. Acquire low-frequency image data and high-frequency image data of a target area through a dual-frequency tomography synthetic aperture radar system, perform neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and perform neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack.

[0059] S2. Based on the principle of dual-frequency tomography, the low-frequency data stack and the high-frequency data stack are fused to obtain a dual-frequency data stack.

[0060] S3. Use a distributed compressed sensing algorithm to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector.

[0061] S4. Generate a three-dimensional point cloud according to the elevation scattering coefficient vector to obtain a three-dimensional image of the target area.

[0062] The dual-frequency tomographic synthetic aperture radar 3D imaging method disclosed in the preferred embodiment of the present invention performs 3D imaging by effectively combining the neighborhood pixel information of single-frequency data with the information contained in dual-frequency data. The dual-frequency tomographic synthetic aperture radar system uses a satellite carrying two radars with wavelengths of λ1 and λ2 to simultaneously acquire dual-band 2D SAR (synthetic aperture radar) image data at the same location. Figure 2 The following is a schematic diagram of the imaging geometry of the dual-frequency tomographic synthetic aperture radar system, where s is the elevation direction, r is the slant range direction, H is the reference satellite's orbital altitude, and θ is the center-down viewing angle of the satellite radar observation geometry. The first pass satellite is set as the reference satellite to calculate the baseline. The baseline length of the nth pass satellite is b. n , after N flights, a set of baselines B is formed.

[0063] The essence of using the tomographic algorithm to solve the target height is to synthesize the elevation scattering coefficient by solving the correlation between the observation vector and the observation matrix. Figure 2 The above diagram shows the resolution unit of the double target overlap at the slant range r. Its elevation scattering coefficient is composed of the ground scatterer and the building top angle. When the dual-frequency data is used to perform three-dimensional information calculation, the following can be obtained: Figure 2 The two vertical scattering coefficient waveforms are shown in medium gray and white curves.

[0064] In a preferred embodiment of the present invention, low-frequency image data and high-frequency image data of the target area are obtained by a dual-frequency tomography synthetic aperture radar system, and the low-frequency image data is subjected to neighborhood pixel joint processing to obtain a low-frequency data stack, and the high-frequency image data is subjected to neighborhood pixel joint processing to obtain a high-frequency data stack. Specifically, for each range-azimuth pixel (x, y) of the low-frequency image data, a preset number of first neighborhood pixels within a preset range of the low-frequency image data are selected to form a low-frequency data stack G(λ1). For the high-frequency image data, the same method is used to select the second neighborhood pixels to construct the high-frequency data stack G(λ2). In a preferred embodiment of the present application, the first neighborhood pixels and the second neighborhood pixels are selected using an 8-pixel neighborhood selection method of vertically adjacent, horizontally adjacent, and diagonally adjacent to obtain 8 neighborhood pixels.

[0065] Furthermore, the low-frequency data stack and the high-frequency data stack are fused to obtain a dual-frequency data stack. Specifically, assuming that there are N coherent observation points in the target area, it is equivalent to forming an equivalent synthetic aperture in the elevation direction. Based on the equivalent synthetic aperture principle, a single-pixel continuous signal model is constructed. The single-pixel continuous signal model is expressed as follows:

[0066]

[0067] Among them, g nrepresents the observation value collected by the nth observation, γ(s) represents the vertical scattering coefficient, ξ n =-2b n / (λr), represents the equivalent spatial frequency of the nth observation, λ represents the radar wavelength, ε n is the noise of the nth observation, and j indicates that this item is the imaginary part of the complex observation value.

[0068] The single-pixel continuous signal model can be regarded as the transformation of the elevation scattering coefficient γ(s) from the spatial time domain to the spatial frequency domain. The single-pixel continuous signal model is discretized to obtain the single-frequency observation signal matrix expression. The single-frequency observation signal matrix expression is:

[0069] g=Rγ+ε

[0070] Among them, R is the observation matrix of size N×L, L is the number of grids divided upward in elevation, and the element in the nth row and lth column of R is R nl =exp(-j2πξ n s l ), s l is the lth pre-divided grid position, ε is the noise vector, and γ represents the elevation scattering coefficient vector.

[0071] When a dual-frequency tomographic synthetic aperture radar system is equipped with two radars for simultaneous acquisition, considering that the target scattering characteristics vary with wavelength, two sets of observation signals can be obtained during the same flight, forming a dual-pixel continuous signal model. The dual-pixel continuous signal model is expressed as:

[0072]

[0073] Here, λ1 represents the radar wavelength for collecting low-frequency image data, and λ2 represents the radar wavelength for collecting high-frequency image data.

[0074] Discretize the dual-pixel continuous signal model to obtain the dual-frequency observation signal matrix expression. The dual-frequency observation signal matrix expression is:

[0075] g(λ)=R(λ)*γ(λ)+ε(λ), λ=λ1orλ2.

[0076] Where R(λ1) represents the observation matrix of the first radar, R(λ2) represents the observation matrix of the second radar, γ(λ1) represents the elevation scattering coefficient vector of the low-frequency image data, γ(λ2) represents the elevation scattering coefficient vector of the high-frequency image data, ε(λ1) represents the noise vector of the low-frequency image data, and ε(λ2) represents the noise vector of the high-frequency image data.

[0077] The pixel values ​​of an image generated by a dual-frequency tomographic synthetic aperture radar system are actually the correlation values ​​between the scattering characteristics of the ground object and the corresponding basis. The scattering characteristics of the ground object are related to the radar wavelength λ. The amplitude and phase characteristics of the ground object in the pixel change with wavelength, but the elevation position remains unchanged. The complex amplitude of the spatial vector synthesized by the two frequency bands will vary due to the amplitude and phase of the data. However, the direction of the spatial vector is only related to the tomographic position of the target area and remains consistent between the two frequency bands. Therefore, the single-frequency tomographic reconstruction algorithm can also be applied to dual-frequency data, and can achieve more accurate calculation of the target area's tomographic position parameters.

[0078] The dual-frequency data obtained by the dual-frequency tomographic synthetic aperture radar system at the same location are equivalent to the same-frequency data obtained at different locations due to different frequency bands. Assume that the actual baseline lengths of a satellite N times are b1, b2, ..., b n The frequency band ratio of the two radars onboard the satellite is α = f2 / f1 = λ1 / λ2, where f1 is the low-frequency band and its corresponding radar wavelength is λ1, which is used to collect low-frequency image data. f2 is the high-frequency band and its corresponding radar wavelength is λ2, which is used to collect high-frequency image data. The equivalent spatial frequencies of the two frequency bands are:

[0079]

[0080] According to the frequency band ratio, the low-frequency image data is equivalent to the sampling point position of the high-frequency image data, and the equivalent spatial frequency of the data combination of the two frequency bands can be obtained:

[0081]

[0082] Based on the combined equivalent spatial frequency, the dual-frequency observation signal matrix expression is subjected to baseline equivalent processing. The low-frequency data stack and the high-frequency data stack are fused using the dual-frequency observation signal matrix expression after baseline equivalent processing to obtain the signal after the single-frequency data neighborhood pixels are combined. The signal expression after the single-frequency data neighborhood pixels are combined is:

[0083]

[0084] Among them, G(λ)=[g1(λ)g2(λ)...g W (λ)] T , ψ(λ)=[R1(λ)R2(λ)...R W (λ)] T , ε(λ)=[ε1(λ)ε2(λ)...ε W (λ)] T, ⊙ represents the Hadamard product of the matrix, g1(λ) represents the observation vector of the first pixel in the low-frequency data stack or the high-frequency data stack, g2(λ) represents the observation vector of the second pixel in the low-frequency data stack or the high-frequency data stack, g W (λ) represents the observation vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack, W represents the number of pixels in the low-frequency data stack or the high-frequency data stack, R1(λ) represents the observation vector of the first pixel in the low-frequency data stack or the high-frequency data stack, R2(λ) represents the observation vector of the second pixel in the low-frequency data stack or the high-frequency data stack, and R W (λ) represents the observation vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack, γ1(λ) represents the elevation scattering coefficient of the first pixel in the low-frequency data stack or the high-frequency data stack, γ2(λ) represents the elevation scattering coefficient of the second pixel in the low-frequency data stack or the high-frequency data stack, and γ W (λ) represents the vertical scattering coefficient of the Wth pixel in the low-frequency data stack or the high-frequency data stack, λ represents the radar wavelength, λ1 represents the radar wavelength for collecting low-frequency image data, λ2 represents the radar wavelength for collecting high-frequency image data, ε1(λ) represents the noise vector of the first pixel in the low-frequency data stack or the high-frequency data stack, ε2(λ) represents the noise vector of the second pixel in the low-frequency data stack or the high-frequency data stack, ε W (λ) represents the noise vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack.

[0085] The signals of the combined single-frequency data neighborhood pixels are spliced ​​together to obtain a dual-frequency data stack, which is expressed as:

[0086]

[0087] Furthermore, a distributed compressed sensing algorithm is used to perform a dual-frequency joint solution on the dual-frequency data stack to obtain the elevation scattering coefficient vector. Specifically, a dual-frequency joint solution model is constructed, and the number of iterations of the dual-frequency joint solution model, the orthogonalized elevation scattering coefficient, the basis vector index set, the dictionary vector set, and the observation vector residual are initialized. The number of iterations is initialized to 1. For each pixel w∈{1,2,...,2W}, the orthogonalized elevation scattering coefficient is Basis vector index set Dictionary vector collection Observation vector residual Represents the observation vector g w The residual after the lth iteration is initialized to

[0088] In the iterative update process, the dual-frequency data stack is input into the initialized dual-frequency joint solution model, the target basis vector that maximizes the sum of the projections of the observation vector residuals is calculated, and the index of the target basis vector is added to the basis vector index set. The basis vector index is expressed as:

[0089]

[0090] in,<a,b> is the inner product of vectors a and b, m l represents the basis vector index selected at the lth iteration, represents the residual of the w-th pixel after the l-1th iteration, R w (λ) represents the observation vector of the w-th pixel. When w∈[1,W], λ=λ1, and when w∈[W,2W], λ=λ2.

[0091] Furthermore, a basis vector is selected by a basis vector index, a dictionary vector is selected from a dictionary vector set, and the selected basis vector is With dictionary vector Perform orthogonalization to obtain the orthogonalized basis vectors And the orthogonalized basis vectors are included in the dictionary vector set. The specific processing process is expressed as:

[0092]

[0093] Wherein, t represents the sequence number when accumulating the orthogonalized values ​​of the vectors selected in the l-1 iterations.

[0094] Furthermore, the orthogonalized elevation scattering coefficient and the observation vector residual are calculated and updated according to the orthogonalized basis vectors. Specifically, the orthogonalized elevation scattering coefficient vector and the observation vector residual are calculated using the orthogonalized basis vectors. The calculation formula of the orthogonalized elevation scattering coefficient vector is:

[0095]

[0096] The calculation formula of the observation vector residual is:

[0097]

[0098] Furthermore, it is determined whether all pixels in the dual-frequency data stack meet the convergence requirement based on the observation vector and the updated observation vector residual. If not, an iterative update process is performed until the convergence requirement is met. In the preferred embodiment of the present application, the convergence judgment rule is: there is no observation vector residual of a pixel in the dual-frequency data stack that is greater than δ times the observation vector, that is, Specifically, if there is a pixel in the dual-frequency data stack with an observation vector residual greater than δ times the observation vector, the iterative update process returns to iteratively calculate the target basis vector to iteratively update the orthogonalized elevation scatter coefficient vector and the observation vector residual. If there is no pixel in the dual-frequency data stack with an observation vector residual greater than δ times the observation vector, indicating that the convergence requirement has been met, orthogonal triangular decomposition is used to deorthogonalize the dictionary vector set containing all orthogonalized basis vectors and the orthogonalized elevation scatter coefficient to obtain the elevation scatter coefficient vector. The parameter δ determines the target error power level that allows the algorithm to converge.

[0099] Specifically, the dictionary vector containing all the orthogonalized basis vectors The only decomposition into an orthogonal matrix Q w and the upper triangular matrix B w , based on the orthogonal matrix Q w and the upper triangular matrix B w , after meeting the convergence requirements, the dictionary vector set and the orthogonalized elevation scattering coefficient are de-orthogonalized to obtain the elevation scattering coefficient vector. The calculation formula of the elevation scattering coefficient vector is:

[0100]

[0101] in, represents the inverse of the upper triangular matrix, Q H represents the transpose of an orthogonal matrix, represents the dictionary vector after meeting the convergence requirements, It represents the orthogonalized vertical scattering coefficient after meeting the convergence requirements.

[0102] Finally, a three-dimensional point cloud is generated according to the elevation scattering coefficient vector to obtain a three-dimensional image of the target area.

[0103] To verify the effectiveness of the dual-frequency tomographic synthetic aperture radar 3D imaging method disclosed in the embodiments of this application, the satellite-borne data of a 1-transmit 20-receive dual-frequency tomographic synthetic aperture radar system was simulated. The specific radar parameters are shown in Table 1.

[0104] Table 1

[0105] Parameter name Parameter information Number of flights 20 X-band frequencies 9.6GHz C-band frequencies 5.6GHz Closest slant distance 663103.97m Total baseline length 783.16m Range resolution 0.92m Azimuth resolution 1.91m Rayleigh resolution 13.23m

[0106] Based on the above parameters, a single scatterer model and a double scatterer model are designed to perform Monte Carlo simulation. For the scene height variation of the single scatterer, three scenes are designed: flat ground, gentle slope and steep slope. The radar cross section (RCS) of the scene is set to follow different Rayleigh distributions, such as Figure 3The following is the image data obtained by the simulated dual-frequency tomographic synthetic aperture radar system. In the single scattering point model, the elevation of the scene is set to 0m~52.4m, as shown in Figure 2. Figure 4 Shown is an elevation image of the single scatterer model.

[0107] Under the same baseline interval, the true height of the scatterer is 0m, and the vertical scattering profile is solved by single frequency solution and dual frequency solution respectively. Figure 5 The following figure shows the vertical scattering profiles for separate imaging of the X-band and C-band using the OMP algorithm (Orthogonal Matching Pursuit) and the DCS algorithm (Distributed Compressed Sensing), respectively. It can be seen that the unambiguous height of the X-band is small, while the unambiguous height of the C-band is large. Therefore, the ambiguity does not appear at the same location, and the peak only appears at the target location. Figure 6 The figure shows the elevation scattering profile imaged using the Dual-OMP algorithm (dual-frequency orthogonal matching pursuit algorithm) and Dual-DCS (dual-frequency tomographic synthetic aperture radar three-dimensional imaging method disclosed in this application). After the X-band data and the C-band data are jointly processed, the peak of the target position is highlighted and the peak of the blurred position is suppressed. Because the tomographic synthetic aperture radar three-dimensional imaging method disclosed in this application combines neighborhood pixel processing, it is less affected by side lobes than the Dual-OMP algorithm, and the target position is estimated more accurately. Therefore, the tomographic synthetic aperture radar three-dimensional imaging method disclosed in this application can effectively highlight the position of the target and greatly improve the accuracy of resolving elevation ambiguity.

[0108] In order to verify the accuracy maintenance capability of the Dual-DCS solution after reducing the number of flybys under different signal-to-noise ratio conditions, a set of single scatterer experiments with different signal-to-noise ratios were designed. The target position was set at 0 meters and the signal-to-noise ratio was set to [1:1:10]dB.

[0109] Figure 7 and Figure 8 The figure shows the experimental results of the single scatterer model. In the figure, OMP-C-20 indicates that the OMP algorithm is used to image the C-band data obtained from 20 flights, OMP-X-20 indicates that the OMP algorithm is used to image the X-band data obtained from 20 flights, Dual-OMP-10 indicates that the Dual-OMP algorithm is used to jointly image the C-band data and X-band data obtained from 10 flights, DCS-C-10 indicates that the DCS algorithm is used to image the C-band data obtained from 10 flights, DCS-X-10 indicates that the DCS algorithm is used to image the X-band data obtained from 10 flights, and Dual-DCS-7 indicates that the Dual-DCS algorithm is used to perform joint neighborhood pixel imaging on the C-band data and X-band data obtained from 7 flights. Figure 7As can be seen, the OMP algorithm's performance with 10 dual-frequency passes is comparable to that of 20 single-frequency passes, effectively halving the number of passes. The DCS algorithm's overall performance is superior to the OMP algorithm because it effectively combines information from eight neighboring pixels, improving the signal-to-noise ratio. Building on the DCS algorithm's ability to halve the number of passes by combining neighboring pixels, the dual-frequency tomographic synthetic aperture radar 3D imaging method disclosed in this application reduces the number of passes to just seven, reducing the number of required passes to one-third.

[0110] In a preferred embodiment of the present invention, a dual-frequency tomographic synthetic aperture radar system is used to acquire low-frequency and high-frequency image data of a target area. Neighborhood pixel joint processing is performed on the low-frequency image data to obtain a low-frequency data stack, and neighborhood pixel joint processing is performed on the high-frequency image data to obtain a high-frequency data stack. Based on the principle of dual-frequency tomography, the low-frequency data stack and the high-frequency data stack are fused to obtain a dual-frequency data stack. A distributed compressed sensing algorithm is used to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector. A three-dimensional point cloud is generated based on the elevation scattering coefficient vector to obtain a three-dimensional image of the target area. The dual-frequency tomographic synthetic aperture radar three-dimensional imaging method provided in this application combines the neighborhood information of single-frequency data to improve the signal-to-noise ratio, thereby reducing the number of required flyovers. It also combines the information contained in the dual-frequency data to effectively highlight the location of scatterers, increase the unambiguous height, and further reduce the number of flyovers, thereby significantly improving the accuracy of resolving elevation ambiguity.

[0111] Accordingly, if Figure 9 As shown, based on a dual-frequency tomographic synthetic aperture radar 3D imaging method, an embodiment of the present invention further provides a dual-frequency tomographic synthetic aperture radar 3D imaging system to implement the dual-frequency tomographic synthetic aperture radar 3D imaging method disclosed in an embodiment of the present invention. The system includes: a data acquisition and processing unit 1, a data fusion unit 2, a dual-frequency joint solution unit 3, and a 3D imaging unit 4;

[0112] The data acquisition and processing unit 1 is configured to acquire low-frequency image data and high-frequency image data of a target area through a dual-frequency tomography synthetic aperture radar system, perform neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and perform neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack;

[0113] The data fusion unit 2 is configured to fuse the low-frequency data stack and the high-frequency data stack based on the dual-frequency tomography principle to obtain a dual-frequency data stack;

[0114] The dual-frequency joint solving unit 3 is used to perform a dual-frequency joint solving on the dual-frequency data stack using a distributed compressed sensing algorithm to obtain an elevation scattering coefficient vector;

[0115] The three-dimensional imaging unit 4 is used to generate a three-dimensional point cloud according to the elevation scattering coefficient vector to obtain a three-dimensional image of the target area.

[0116] For the specific definition of a dual-frequency tomographic synthetic aperture radar three-dimensional imaging system, please refer to the above-mentioned definition of a dual-frequency tomographic synthetic aperture radar three-dimensional imaging method, which will not be repeated here. Those skilled in the art will appreciate that the modules and steps described in conjunction with the embodiments disclosed in the present invention can be implemented in hardware, software, or a combination of both. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0117] like Figure 10 As shown, an embodiment of the present invention provides a computer device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps in the above-mentioned dual-frequency tomography synthetic aperture radar three-dimensional imaging method embodiment are implemented, for example Figure 1 Steps S1 to S4 described in .

[0118] Those skilled in the art will understand that the schematic Figure 10 These are merely examples of computer devices and do not constitute limitations on the computer device. The computer device may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0119] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the computer device, connecting various parts of the entire computer device using various interfaces and lines.

[0120] The memory can be used to store the computer programs and / or modules, and the processor implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created based on the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0121] Wherein, if the module integrated in the computer device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0122] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0123] Accordingly, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device containing the computer-readable storage medium is controlled to execute the steps of the dual-frequency tomography synthetic aperture radar three-dimensional imaging method as described in the above embodiment, for example Figure 1 Steps S1 to S4 described in .

[0124] The present embodiment provides a tomographic synthetic aperture radar (TAR) 3D imaging method, system, device, and medium for addressing the technical issues of high flyover times and low 3D imaging accuracy associated with existing TMR 3D imaging methods. A dual-frequency TMR system acquires low-frequency and high-frequency image data of a target area. Neighborhood pixel joint processing is performed on the low-frequency image data to obtain a low-frequency data stack, and neighborhood pixel joint processing is performed on the high-frequency image data to obtain a high-frequency data stack. Based on the principle of dual-frequency tomography, the low-frequency and high-frequency data stacks are fused to obtain a dual-frequency data stack. A distributed compressed sensing algorithm is used to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector. A 3D point cloud is generated based on the elevation scattering coefficient vector to obtain a 3D image of the target area. The dual-frequency TMR 3D imaging method provided herein combines neighborhood information from single-frequency data to improve the signal-to-noise ratio, thereby reducing the number of required flyovers. The information contained in the dual-frequency data is combined to effectively highlight the location of scatterers, increasing the unambiguous height while further reducing the number of flyovers, significantly improving the accuracy of elevation ambiguity resolution.

[0125] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the various technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] The above-described embodiments merely represent several preferred embodiments of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that a person skilled in the art could make several improvements and substitutions without departing from the technical principles of the present invention, and such improvements and substitutions should also be considered within the scope of the present invention. Therefore, the scope of the present invention should be determined by the scope of the claims.

Claims

1. A dual-frequency tomographic synthetic aperture radar three-dimensional imaging method, characterized in that: The method comprises: Acquire low-frequency image data and high-frequency image data of the target area through a dual-frequency tomographic synthetic aperture radar system, perform neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and perform neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack; Based on the principle of dual-frequency tomography, the low-frequency data stack and the high-frequency data stack are fused to obtain a dual-frequency data stack; A distributed compressed sensing algorithm is used to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector; A three-dimensional point cloud is generated according to the elevation scattering coefficient vector to obtain a three-dimensional image of the target area.

2. The dual-frequency tomographic synthetic aperture radar three-dimensional imaging method according to claim 1, wherein: The performing neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and performing neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack, comprises: In the low-frequency image data, selecting a preset number of first neighboring pixels within a preset range for each range-azimuth pixel to construct a low-frequency data stack; In the high-frequency image data, selecting a preset number of second neighboring pixels within the preset range for each range-azimuth pixel to construct a high-frequency data stack; The first neighborhood pixels and the second neighborhood pixels both include vertically adjacent pixels, horizontally adjacent pixels, and diagonally adjacent pixels.

3. The dual-frequency tomographic synthetic aperture radar three-dimensional imaging method according to claim 1, wherein: The method of fusing the low-frequency data stack and the high-frequency data stack based on the dual-frequency tomography principle to obtain a dual-frequency data stack includes: Based on the equivalent synthetic aperture principle, a single-pixel continuous signal model is constructed, and the single-pixel continuous signal model is expanded into a dual-pixel continuous signal model; Discretizing the dual-pixel continuous signal model to obtain a dual-frequency observation signal matrix expression; Performing baseline equivalent processing on the dual-frequency observation signal matrix expression; The dual-frequency observation signal matrix expression after baseline equivalent processing is used to fuse the low-frequency data stack and the high-frequency data stack to obtain a dual-frequency data stack.

4. The dual-frequency tomographic synthetic aperture radar three-dimensional imaging method according to claim 1, wherein: The dual-frequency data stack is represented as: Where g1(λ) represents the observation vector of the first pixel in the low-frequency data stack or the high-frequency data stack, g2(λ) represents the observation vector of the second pixel in the low-frequency data stack or the high-frequency data stack, and g W (λ) represents the observation vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack, W represents the number of pixels in the low-frequency data stack or the high-frequency data stack, R1(λ) represents the observation vector of the first pixel in the low-frequency data stack or the high-frequency data stack, R2(λ) represents the observation vector of the second pixel in the low-frequency data stack or the high-frequency data stack, and R W (λ) represents the observation vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack, ⊙ represents the Hadamard product of the matrix, γ1(λ) represents the elevation scattering coefficient of the first pixel in the low-frequency data stack or the high-frequency data stack, γ2(λ) represents the elevation scattering coefficient of the second pixel in the low-frequency data stack or the high-frequency data stack, and γ W (λ) represents the vertical scattering coefficient of the Wth pixel in the low-frequency data stack or the high-frequency data stack, λ represents the radar wavelength, λ1 represents the radar wavelength for collecting low-frequency image data, λ2 represents the radar wavelength for collecting high-frequency image data, ε1(λ) represents the noise vector of the first pixel in the low-frequency data stack or the high-frequency data stack, ε2(λ) represents the noise vector of the second pixel in the low-frequency data stack or the high-frequency data stack, ε W (λ) represents the noise vector of the Wth pixel in the low-frequency data stack or the high-frequency data stack.

5. The dual-frequency tomographic synthetic aperture radar three-dimensional imaging method according to claim 4, wherein: The method of using a distributed compressed sensing algorithm to perform a dual-frequency joint solution on the dual-frequency data stack to obtain an elevation scattering coefficient vector includes: Constructing a dual-frequency joint solution model, initializing the number of iterations of the dual-frequency joint solution model, the orthogonalized vertical scattering coefficient, the basis vector index set, the dictionary vector set and the observation vector residual; In an iterative update process, the dual-frequency data stack is input into the initialized dual-frequency joint solution model, a target basis vector that maximizes the sum of the projections of the observation vector residuals is calculated, and an index of the target basis vector is added to the basis vector index set; Selecting a basis vector through the basis vector index set, selecting a dictionary vector from the dictionary vector set, orthogonalizing the selected basis vector and the dictionary vector to obtain an orthogonalized basis vector, and incorporating the orthogonalized basis vector into the dictionary vector set; Calculate and update the orthogonalized vertical scatter coefficient and the observation vector residual according to the orthogonalized basis vector; determining whether all pixels of the dual-frequency data stack meet the convergence requirement based on the observation vector and the updated observation vector residual, and if not, performing the iterative update process until the convergence requirement is met; Orthogonal triangular decomposition is used to de-orthogonalize the dictionary vector set and the orthogonalized elevation scattering coefficient after the convergence requirement is met, so as to obtain an elevation scattering coefficient vector.

6. The dual-frequency tomographic synthetic aperture radar three-dimensional imaging method according to claim 5, wherein: The orthogonal triangular decomposition is used to deorthogonalize the dictionary vector set and the orthogonalized elevation scattering coefficient after meeting the convergence requirement to obtain the elevation scattering coefficient vector, including: Uniquely decomposing the dictionary vector after meeting the convergence requirement into an orthogonal matrix and an upper triangular matrix; Based on the orthogonal matrix and the upper triangular matrix, de-orthogonalization processing is performed on the dictionary vector set and the orthogonalized elevation scattering coefficient that meet the convergence requirement to obtain an elevation scattering coefficient vector.

7. The dual-frequency tomographic synthetic aperture radar three-dimensional imaging method according to claim 6, wherein: The calculation expression of the elevation scattering coefficient vector is: in, represents the inverse of the upper triangular matrix, Q H represents the transpose of an orthogonal matrix, represents the dictionary vector after meeting the convergence requirements, It represents the orthogonalized vertical scattering coefficient after meeting the convergence requirements.

8. A dual-frequency tomographic synthetic aperture radar 3D imaging system, implementing the dual-frequency tomographic synthetic aperture radar 3D imaging method according to any one of claims 1 to 7, characterized in that: The system includes: a data acquisition and processing unit, a data fusion unit, a dual-frequency joint solution unit and a three-dimensional imaging unit; The data acquisition and processing unit is configured to acquire low-frequency image data and high-frequency image data of the target area through a dual-frequency tomography synthetic aperture radar system, perform neighborhood pixel joint processing on the low-frequency image data to obtain a low-frequency data stack, and perform neighborhood pixel joint processing on the high-frequency image data to obtain a high-frequency data stack; The data fusion unit is configured to fuse the low-frequency data stack and the high-frequency data stack based on the dual-frequency tomography principle to obtain a dual-frequency data stack; The dual-frequency joint solving unit is used to perform a dual-frequency joint solution on the dual-frequency data stack using a distributed compressed sensing algorithm to obtain an elevation scattering coefficient vector; The three-dimensional imaging unit is used to generate a three-dimensional point cloud according to the elevation scattering coefficient vector to obtain a three-dimensional image of the target area.

9. A computer device, characterized in that: The computer device includes a memory, a processor, and a transceiver, which are connected via a bus; the memory is used to store a set of computer program instructions and data, and transmit the stored data to the processor, and the processor executes the program instructions stored in the memory to perform the dual-frequency tomography synthetic aperture radar three-dimensional imaging method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the dual-frequency tomography synthetic aperture radar three-dimensional imaging method according to any one of claims 1 to 7 is implemented.

Images