SAR (Synthetic Aperture Radar) imaging method based on mixed norm depth expansion network and related device

Abstract

The invention discloses an SAR imaging method based on a mixed norm depth expansion network and a related device, and relates to the technical field of remote sensing imaging, and the method comprises the steps: obtaining radar working parameters; obtaining an echo matrix according to the SAR original echo signals collected at all sampling moments; determining an observation scene of the SAR, and constructing a corresponding slant distance matrix on each scene point in the observation scene at each random sampling moment; determining an echo signal after range migration correction according to the echo matrix and the slant distance matrix; and inputting the echo signal after range migration correction into the trained mixed norm depth expansion network to obtain a reconstruction scene, the optimization equation is converted into the mixed norm optimization equation based on the alternating direction multiplier method, the mixed norm optimization equation comprises the Tikhonov regularization optimization equation and the L1 / 2 regularization optimization equation, and the reconstruction scene is obtained. And a network approximate iteration process is deeply developed, and the calculation complexity of the iteration process is reduced.

Description

Technical Field

[0001] The present application relates to the field of remote sensing imaging technology, and in particular to a SAR imaging method and related devices based on a mixed norm deep expansion network. Background Art

[0002] Spaceborne synthetic aperture radar (SAR) can efficiently detect hot spots and key targets through multi-level situational awareness techniques to achieve rapid discovery, detection and identification of scene targets. With the development of compressed sensing (CS) theory, under the condition that the observation matrix satisfies the restricted isometry property (RIP), the signal can be recovered from a non-uniform sub-Nyquist sampling method with far fewer Nyquist sampling samples based on the sparsity of the signal. Under the condition that the nonlinear iterative optimization algorithm guarantees high spatial resolution, the compressed sensing radar can achieve ultra-wide-area observation of the scene through a smart beam pointing design. The compressed sensing radar can increase the width / resolution ratio to the same order of magnitude as the SIMO-SAR and MIMO-SAR systems. The compressed sensing radar system based on sub-Nyquist sampling has a simple structure, small antenna size, and small data volume, which reduces the hardware burden, signal storage and transmission pressure. It is a very attractive and promising research route for high-resolution wide-band SAR systems.

[0003] Compressed sensing radar systems require prior information on the sparsity of the observed scene, which has certain limitations in application. For example, in most cases, the sparsity of the observed scene is unknown, and CS theory is applied to complex scenes. At present, sparse imaging algorithms mainly include greedy algorithms, algorithms based on statistical distribution, and regularization algorithms. Greedy algorithms tend to fall into local optimality. Algorithms based on statistical distribution have powerful modeling capabilities, but they have certain challenges for complex scenes that lack sparsity. In regularization algorithms, although there are some useful heuristic methods, there is no unified standard for how to choose appropriate regularization parameters. Parameters can only be selected based on experience and manually tuned by "trial and error", which has low execution efficiency and long running time. At the same time, due to the limitations of the algorithm itself, the compressed sensing imaging method can only obtain less image information and cannot obtain rich deep image information, which results in poor reconstruction performance of the compressed sensing algorithm when the undersampling multiple is large. Therefore, a fast reconstruction algorithm with reliable accuracy and close to ideal theoretical performance is also a difficult problem in CS research. Summary of the invention

[0004] The purpose of this application is to provide a SAR imaging method and related devices based on a hybrid norm deep expansion network, which, under low signal-to-noise ratio conditions, takes into account the impact of noise, clutter, etc. on the algorithm, enhances the stability of the imaging process against interference, and is not affected by scene inhomogeneity.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a SAR imaging method based on a mixed norm deep unfolding network, comprising:

[0007] Acquire radar operating parameters; the radar operating parameters include scene center reference slant distance, satellite flight speed, azimuth antenna size, number of transmitted pulses, modulation frequency of transmitted signals, time vector of radar transmitted pulses and operating wavelength;

[0008] Acquire the echo matrix according to the SAR original echo signals collected at all sampling moments;

[0009] Determine the SAR observation scene and construct the corresponding slant range matrix of each scene point in the observation scene at each random sampling moment;

[0010] Determine the echo signal after range migration correction according to the echo matrix and the slant range matrix;

[0011] The echo signal after the range migration correction is input into the trained hybrid norm deep expansion network to obtain a reconstructed scene; the hybrid norm deep expansion network includes K levels, each level includes four submodules, namely a first regularization module, a sparse transformation module, a second regularization module and a multiplier update module, and the reconstruction module outputs the result; in any level, the data input in the previous level is used to obtain a reconstructed scene through the reconstruction module, the first regularization module is used to solve the Tikhonov regularization equation to obtain an iterative solution result; the sparse representation of the iterative solution result is obtained through the sparse transformation module; the second regularization module is used to solve L 1 / 2 Regularize the equation and obtain the iterative solution result; the multiplier update module is used to implement the update of the Lagrange multiplier.

[0012] In a second aspect, the present application provides a SAR imaging device based on a mixed norm deep expansion network, comprising:

[0013] The radar working parameter acquisition module is used to: acquire radar working parameters; the radar working parameters include scene center reference slant distance, satellite flight speed, azimuth antenna size, number of transmitted pulses, modulation frequency of transmitted signal, time vector of radar transmitting pulse and working wavelength;

[0014] The echo matrix acquisition module is used to: acquire the echo matrix according to the SAR original echo signals collected at all sampling moments;

[0015] The slant range matrix construction module is used to: determine the SAR observation scene and construct the corresponding slant range matrix of each scene point in the observation scene at each random sampling moment;

[0016] A module for determining an echo signal after range migration correction, used to: determine an echo signal after range migration correction according to the echo matrix and the slant range matrix;

[0017] The scene reconstruction module is used to: input the echo signal after the range migration correction into the trained hybrid norm deep expansion network to obtain a reconstructed scene; the hybrid norm deep expansion network includes K levels, each level includes four submodules, namely a first regularization module, a sparse transformation module, a second regularization module and a multiplier update module, and the reconstruction module outputs the result; in any level, the data input in the previous level is used to obtain a reconstructed scene through the reconstruction module, the first regularization module is used to solve the Tikhonov regularization equation to obtain an iterative solution result; the sparse representation of the iterative solution result is obtained through the sparse transformation module; the second regularization module is used to solve L 1 / 2 Regularize the equation and obtain the iterative solution result; the multiplier update module is used to implement the update of the Lagrange multiplier.

[0018] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned SAR imaging method based on a mixed norm deep expansion network.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned SAR imaging method based on a mixed norm deep expansion network.

[0020] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0021] The existing compressed sensing algorithm does not take into account the influence of noise and clutter under low signal-to-noise ratio conditions. In addition, considering the inhomogeneity of the scene, a single regularization parameter weighting cannot distinguish between strong and weak targets well. Based on this, the present application provides a SAR imaging method and related devices based on a mixed norm deep expansion network. Taking into account the inhomogeneity of the scene, a single regularization parameter weighting cannot distinguish between strong and weak targets well. Taking into account sparsity, stability, and dynamic parameter adaptability, the optimization equation is converted into a mixed norm optimization equation based on the alternating direction multiplier method. The mixed norm optimization equation includes: solving the Tikhonov regularization optimization equation through the first regularization module in the mixed norm deep expansion network, and the global optimum can be obtained by derivation; L is regularized by the second regularization module in the mixed norm deep expansion network. 1 / 2 Regularized optimization equations are solved analytically. A deep unfolding network is also used to approximate the iterative process, reducing the computational complexity of the iterative process and further optimizing the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 This is an application environment diagram of a SAR imaging method based on a mixed norm deep expansion network in an embodiment of the present application;

[0024] Figure 2 A schematic flow chart of a SAR imaging method based on a mixed norm deep expansion network provided in one embodiment of the present application;

[0025] Figure 3 A schematic diagram of functional modules of a SAR imaging device based on a hybrid norm deep expansion network provided in one embodiment of the present application;

[0026] Figure 4 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0028] The compressed sensing radar system has a simple structure, small antenna size, and small data volume, which reduces the hardware burden, signal storage and transmission pressure, and is a very attractive and promising research route for high-resolution wide-band SAR systems. Due to the limitations of the algorithm itself, the compressed sensing radar can only obtain less image information and cannot obtain rich deep image information, which results in poor reconstruction performance of the compressed sensing algorithm when the undersampling multiple is large.

[0029] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0030] The SAR imaging method based on the mixed norm deep expansion network provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the radar operating parameters to the server 104. After the server 104 receives the radar operating parameters, for the radar operating parameters, the server 104 obtains the echo matrix based on the SAR original echo signals collected at all sampling times, determines the SAR observation scene, and constructs the corresponding slant range matrix at each scene point in the observation scene at each random sampling time, determines the echo signal after the range migration correction based on the echo matrix and the slant range matrix, and inputs the echo signal after the range migration correction into the trained hybrid norm deep expansion network to obtain the reconstructed scene. The server 104 can feedback the obtained reconstructed scene to the terminal 102. In addition, in some embodiments, the SAR imaging method based on the mixed norm deep expansion network can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform SAR imaging processing on the radar operating parameters, or the server 104 can obtain the radar operating parameters from the data storage system and perform SAR imaging processing on the radar operating parameters.

[0031] The terminal 102 may be, but is not limited to, various desktop computers and laptop computers. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0032] In an exemplary embodiment, Figure 2 As shown, a SAR imaging method based on a hybrid norm deep expansion network is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 201 to 205. Among them:

[0033] Step 201: Acquire radar operating parameters; the radar operating parameters include scene center reference slant distance, satellite flight speed, azimuth antenna size, number of transmitted pulses, modulation frequency of transmitted signal, time vector of radar transmitted pulse and operating wavelength.

[0034] Step 202: Acquire an echo matrix according to the SAR original echo signals collected at all sampling moments.

[0035] Step 203: Determine the observation scene of the SAR, and construct the slant range matrix corresponding to each scene point in the observation scene at each random sampling moment.

[0036] Step 204: Determine the echo signal after range migration correction according to the echo matrix and the slant range matrix.

[0037] Step 205: Input the echo signal after the range migration correction into the trained hybrid norm deep expansion network to obtain a reconstructed scene; the hybrid norm deep expansion network includes K levels, each level includes four submodules, namely a first regularization module, a sparse transformation module, a second regularization module and a multiplier update module, and the reconstruction module outputs the result; in any level, the data input in the previous level is used to obtain a reconstructed scene through the reconstruction module, the first regularization module is used to solve the Tikhonov regularization equation to obtain an iterative solution result; the sparse representation of the iterative solution result is obtained through the sparse transformation module; the second regularization module is used to solve L 1 / 2 Regularize the equation and obtain the iterative solution result; the multiplier update module is used to implement the update of the Lagrange multiplier.

[0038] In the implementation of the above steps 201 to 205, considering the non-uniformity of the scene, a single regularization parameter weighting cannot well distinguish between strong and weak targets, and comprehensively considering sparsity, stability, and dynamic parameter adaptability, the optimization equation is transformed into a hybrid norm optimization equation based on the alternating direction multiplier method. The hybrid norm optimization equation includes: solving the Tikhonov regularization optimization equation through the first regularization module in the hybrid norm deep expansion network, and the global optimum can be obtained by derivation; L is regularized by the second regularization module in the hybrid norm deep expansion network. 1 / 2 Regularized optimization equations are solved analytically. A deep unfolding network is also used to approximate the iterative process, reducing the computational complexity of the iterative process and further optimizing the algorithm.

[0039] In step 201, the radar operating parameters are f1 = {R0, V e ,L a ,Na ,λ,Kr,ta}, where R0 is the reference slant distance from the center of the scene, V e is the satellite flight speed, L a is the antenna size in azimuth, N a is the number of transmitted pulses, λ is the working wavelength, Kr is the modulation frequency of the transmitted signal, and the time vector of the radar transmitting pulse ta=[η1,η2,...,η m ,...η Na ] T , where η m represents the non-uniform sampling time when the SAR transmits the mth pulse, [·] T Represents a transpose operation.

[0040] In another exemplary embodiment of the present application, the echo pixel points of the SAR are a two-dimensional matrix, and the echo matrix in step 202 is expressed as:

[0041]

[0042] Among them, S echo,n,k Indicates that the sampling time is η m , the center slope distance is τ k The echo of the target point c / 2, that is, the echo matrix S echo The element in the mth row and kth column. In addition, for the convenience of representation, S echo Also represented as a column vector S of element row vectors echo =[S echo,1,col … S echo,m,cZl … S echo,Na,col ] T , and a row vector S whose elements are column vectors echo =[S echo,row,1 … S echo,row,k … S echo,row,Nr ]. Where S echo,n,col =[S echo,m,1 … S echo,m,k …S echo,m,Nr ] indicates S echo The mth row, S echo,row,k =[S echo,1,k … S echo,m,k … S echo,Na,k ] T Indicates S echo The kth column of .

[0043] The specific process of step 203 is: select a scene, establish a geometric motion model between the SAR system and the ground scene (i.e., the observation scene), and construct the slant range matrix r (η m ).

[0044] Taking the center point of the reference scene as the center, take an area not smaller than the size of the observation scene and divide the area into grids. Assume that there are N points along the x-axis and N points along the y-axis. The spacing between grids should be less than or equal to the azimuth resolution ρ. The spacing between grids is recorded as l, l≤ρ a However, the grid spacing l should not be too small, because too small a grid spacing cannot optimize the reconstruction performance, but will affect the efficiency of the algorithm. i ,y i ,0), the coordinates of each scene point can be obtained as shown in the following formula (2):

[0045]

[0046] Assume that the initial position of the satellite sampling is [x track ,0,H], then the satellite's sequential sampling position coordinates are {[x track +V e η1,0,H],[x track +Vη2,0,H],…,[x track +Vη m ,0,H],…[x track +Vη Na ,0,H]}, where η1 is the zero time of random sampling and H is the flight altitude of the satellite. Finally, the slant distance r(η) of all scene points at each sampling time is calculated based on the scene point coordinates and the satellite sampling position coordinates. m ).

[0047] In the present invention, any slope distance r is calculated in the non-rotating geocentric coordinate system. i (η m ), while the r(η m ) The time difference between the periapsis and the SAR system startup time needs to be considered. i (η m ) is calculated as follows:

[0048] First, the slope distance r in the geocentric coordinate system is calculated using formula (3); formula (3) is expressed as follows:

[0049]

[0050] Among them, a is the semi-major axis of the elliptical orbit, and M is the mean pericentric angle.

[0051] Secondly, according to the slant distance r obtained above, the satellite sampling position coordinates are converted to the non-rotating geocentric coordinate system according to the following formula:

[0052]

[0053]

[0054]

[0055]

[0056] The (x, y, z) in the last formula (7) is the coordinate of the satellite sampling position in the non-rotating geocentric coordinate system. In the above formula, (ξηξ) represents the coordinate of a point on the elliptical orbit in the reference system ξηξ. θ represents the true pericentric angle, ω represents the pericentric point distance, Ω represents the intersection radius, i represents the orbital inclination, and l1, l2, l3, m1, m2, m3, n1, n2, and n3 are all intermediate parameters.

[0057] Next, according to the satellite sampling position coordinates and scene point coordinates in the non-rotating geocentric coordinate system, the slant distance r (η m ).

[0058] At the random sampling moment in each azimuth, the slant range coefficients of all scene points at the random sampling moment are calculated. The slant range coefficients are two-dimensional matrix r, which are specifically expressed as follows:

[0059]

[0060] in, Denotes the crossing time as η m , set the slope range coefficient of the (n,n)th target in the scene target, that is, the mth row and nth row in the slope range coefficient matrix r 2 In addition, for the convenience of representation, r n Also represented as a column vector of row vectors and a row vector r=[r ta,1,1 … r ta,n,n … r ta,N,N ].in represents the mth row of r, Indicates the nth 2 List.

[0061] In another exemplary embodiment of the present application, the above step 204 can be replaced by the following steps 301 to 303:

[0062] Step 301: Determine a scene delay signal matrix according to the echo matrix and the slant range matrix;

[0063] Step 302: using projection technology to perform signal enhancement on the scene delay signal matrix to obtain a projected scene delay signal matrix;

[0064] Step 303: summing and superimposing the projected scene delay signal matrix to determine the echo signal after range migration correction.

[0065] In another exemplary embodiment of the present application, step 301 may include the following steps 401 to 406:

[0066] Step 401: performing a range-to-Fourier transform on the echo matrix to obtain an echo signal after the range-to-Fourier transform;

[0067] Step 402: performing range compression matched filtering on the echo signal after the range Fourier transform to obtain filtered echo information;

[0068] Step 403: performing a range-direction inverse Fourier transform on the filtered echo information to obtain a range-compressed echo signal;

[0069] Step 404: Based on the slant distance matrix constructed in step 203, use the formula t=r(η m ) / 2c, determine the time delay from SAR transmitting signal to receiving reflected signal; where r(η m ) is at each sampling time η m The corresponding slant distance matrix of each scene point in the observation scene, c is the speed of light;

[0070] Step 405: Find the echo signals of each scene point corresponding to the time delay from the echo signals after distance compression;

[0071] Step 406: Construct a Na×N 2 The first empty matrix, and the echo signals of each scene point corresponding to the delay are placed in Na×N 2 The first empty matrix of the scene delay signal matrix is ​​obtained; where N a is the number of transmitted pulses, and N is the number of scene points along the x-axis or y-axis of the observed scene.

[0072] The specific implementation process of steps 401 to 403 is as follows:

[0073] (1) First, the echo matrix S echo Performing distance Fourier transform, the Fourier transform of the embodiment of the present invention is different from the fast Fourier transform (FFT), and the specific operation method is as follows:

[0074] Extract S echo The first row of this matrix is r / 2 elements and N r / 2 elements are swapped, and then FFT is performed on the swapped row matrix. After the FFT operation is completed, the first half of the elements and the second half of the elements of the row matrix after FFT are swapped again, and the distance Fourier transform of the first row of echo signal data is completed. The above operation is performed on each row of the echo matrix in turn to complete the distance Fourier transform of the original echo signal matrix, and the signal after the distance Fourier transform is recorded as S echo,r_fft .

[0075] (2) Then construct a matched filter to achieve distance compression. The filter expression is as follows:

[0076]

[0077] Convert the distance to the Fourier transformed signal S echo,r_fft Each row of H rc Multiply point by point, and record the multiplied two-dimensional matrix as S echo,rc_fft Then, S echo,rc_fft The distance is padded with α times of zero at both ends of the dimension. The matrix after zero padding is Then perform the distance Fourier inverse transform to complete the pulse compression, and the signal after the distance compression is recorded as S echo,rc , S echo,rc This is the output parameter of this module. The method of distance to inverse Fourier transform is the same as the steps of distance to Fourier transform, except that the FFT function used in the process is replaced by the IFFT function.

[0078] In step 302, the specific process of obtaining the projected scene delay signal matrix is ​​as follows:

[0079] First, construct the projection matrix D (n,n) for Where λ is the operating wavelength, and At random sampling time η1, η m and η Na The slant range coefficient of the (n,n)th target in the observation scene, [·] T Represents a transpose operation.

[0080] Then, according to the formula S echo,projection (n,n)=D (n,n )·([D (n,n) ] H ·S echo,interp (n,n)), projecting the scene delay signal matrix; where S echo,projection (n,n) is the nth 2 Column scene delay signal, [] H represents the conjugate transpose operation, Secho,interp (n,n) is the nth in the scene delay signal matrix 2 Column scene delay signal.

[0081] Finally, construct a Na×N 2 The second empty matrix and the projected scene delay signal is placed in Na×N 2 The second empty matrix is ​​used to obtain the scene delay signal matrix after projection.

[0082] The specific process of step 303 is: construct a Na×N empty matrix, denoted as S echo,rcm At each sampling time η m , sum and superimpose the scene delay signals on the same range gate and put them into S echo,rcm For example, in the first range gate, that is, when n=1, the data of all scene points arranged on n=1 are taken out, summed and added into S echo,rcm In the first column of the mth row, when it is the second range gate, the data of all scene points arranged on n = 2 are taken out, summed and superimposed and put into S echo,rcm In the second column of the mth row, and so on, the range migration correction is completed. The signal after range migration correction is recorded as S echo,rcm , S echo,rcm This is the output parameter of this module.

[0083] Before inputting the echo signal after the range migration correction into the trained hybrid norm deep expansion network, the SAR imaging method based on the hybrid norm deep expansion network also includes: acquiring a training data set; the training data set includes a number of observation scenes and the echo corresponding to each observation scene; using the training data set to train the hybrid norm deep expansion network to obtain a trained hybrid norm deep expansion network.

[0084] The training data set is constructed based on the radar working parameters. The process is as follows:

[0085] The reconstruction matrix is ​​constructed based on the slant range of the demarcated scene, which is expressed as follows:

[0086]

[0087] The two-dimensional matrix D is a reconstruction matrix obtained according to the slant range of the demarcated scene; where D m,n =exp{-j4πR n (η m ) / λ}·ε(((acos(R0 / R n (η m ))-0.886*λ / L a ))) indicates that the crossing time is η m , slope distance is R n (ηm ) is the reconstruction element of the target point, ε(·) represents the step function, c is the speed of light, that is, the element in the mth row and nth column of the reconstruction matrix D.

[0088] The corresponding parameter pairs are the training data sets of the mixed norm deep expansion network (compressed sensing radar super-resolution imaging network). in P is the total amount of data in the training set. is the observation scene corresponding to the pth group of radar operating parameters, for The corresponding echo.

[0089] Next, we build the network structure, perform network mapping, initialize network parameters and train the network.

[0090] The hybrid norm deep expansion network includes K levels, each level contains four sub-modules, namely the first regularization module, the sparse transformation module, the second regularization module and the multiplier update module, and the reconstruction module outputs the result.

[0091] Step 1, network mapping:

[0092] 1) The first regularization module z (k) By solving the Tikhonov regularization equation, an iterative solution result of the first regularization module is obtained.

[0093] The Tikhonov regularization equation is expressed as shown in formula (10):

[0094]

[0095] Among them, λ and ρ are regularization parameters, and z is the solution sought by regularization.

[0096] The iterative solution results of the first regularization module are expressed as follows:

[0097]

[0098] Among them, z (k) represents the k-th iteration solution result of the first regularization module, k represents the number of iterations, ρ (k) and λ (k) represents the optimization parameter of the current k-th iteration, σ (k-1) and (k-1) They respectively represent the iterative update results of the observation scene σ and the Lagrange multiplier y of the k-1th iteration of the previous layer.

[0099] 2) Sparse transformation module Γ(·) (k) :The sparse transformation module obtains a sparse solution through a nonlinear sparse transformation domain Γ(·).

[0100] Γ(·) (k) represents the sparse transform domain of the scene at the kth layer. For the sparse transform domain Γ(·) (k) In order to enhance the network representation capability and universal approximation characteristics, this application constructs a sparse transformation module through a convolutional neural network (CNN) and designs a universal nonlinear sparse transformation function. In this application, the sparse transformation module consists of three convolutional layers (without bias setting). Among them, the first convolutional layer and the third convolutional layer each correspond to 1 filter, the size of which is 3×3. The second convolutional layer corresponds to 32 filters, the size of which is 3×3. Mathematically, the sparse transformation is reversible, and the symmetric network Γ -1 (·) (k) Represented by a mirror-symmetric frame, Γ(·) (k) ·Γ -1 (·) (k) = Ι, Ι is the unit matrix.

[0101] In addition, since the reconstruction of the radar system scene is a complex operation, a model with real and imaginary parts separated (real-complex model) is established, which is specifically expressed as follows:

[0102]

[0103]

[0104] Where D mn The crossing time is η m , slope distance is R n (η m )The reconstruction element of the target point, (D mn ) real is the real part of the reconstructed element, (D mn ) imag is the imaginary part of the reconstructed element; (z (k) ) real , (z (k) ) imag They are the real and imaginary parts of the k-th iterative solution of the first regularization module.

[0105] According to the above real-complex model, it is divided into two channels for training and learning.

[0106] 3) The second regularization module σ (k) :The second regularization module solves L 1 / 2 norm equation to achieve good reconstruction performance, which is beneficial to the recovery and reconstruction of low signal-to-noise ratio targets.

[0107] L 1 / 2 The norm equation is expressed as shown in formula (14):

[0108]

[0109] In the above formula, s represents the echo signal S after the range migration correction in step 303. echo,rcm , D is the reconstruction matrix obtained according to the slant range of the demarcated scene in formula (10).

[0110] The iterative solution results of the second regularization module are expressed as follows:

[0111]

[0112] in,

[0113]

[0114]

[0115]

[0116]

[0117] In the above formula, σ (k) represents the k-th iteration solution result of the second regularization module, is the intermediate variable, u (k) represents an optimized iteration parameter in the kth iteration process, D H represents the conjugate transpose of the reconstruction matrix D, represents the result of the backscattering coefficient of the i-th scene during the k-th iteration, is the intermediate variable, is the intermediate variable, represents the arccosine function, usually taken as ρ (k) is the coefficient of the quadratic penalty term.

[0118] 4) Multiplier update module y (k) : This multiplier update module is used to update the Lagrange multiplier y (k) .

[0119] y (k) =y (k-1) +τ (k) ρ (k) (z k -σ k ) (twenty one).

[0120] Step 2, initialization.

[0121] Network input initialization: in, Initial input to the mixture norm deep unrolled network.

[0122] Network parameter initialization: τ (0) =1,ρ (0) =0.5, Among them, τ (0) is the initial iteration step length, ρ (0) and λ (0) is the coefficient of the initial quadratic penalty term.

[0123] The network parameters of each layer of the mixed norm deep expansion network are learned through end-to-end training instead of being fixed.

[0124] Step 3, based on the data set Training Mixture Norm Deep Unrolled Networks.

[0125] In this application, reconstruction error and sparsity constraints are considered in the loss function, and the optimization parameter κ is used to control the balance between the two constraints. The loss function is expressed as follows:

[0126]

[0127] Among them, Loss{} represents the loss function, τ (k) is the iteration step, σ is the observation scene, is the reconstructed scene corresponding to the observed scene, K is the number of network layers of the mixed norm deep expansion network, and κ represents the regularization parameter used to balance the first 2-norm term and the second 2-norm term.

[0128] After determining the loss function, the parameters can be learned autonomously to obtain a trained mixed norm deep expansion network.

[0129] After obtaining the trained mixed norm deep expansion network, the echo signal S after range migration correction is echo,rcm , load it into the trained mixed norm deep expansion network, and you can get the desired reconstructed scene.

[0130] The present application also provides an application scenario, which applies the above-mentioned SAR imaging method based on the mixed norm deep expansion network. Specifically: the SAR imaging method based on the mixed norm deep expansion network provided in this embodiment can be applied in the SAR imaging scenario. The SAR imaging scenario includes a parameter acquisition link and a SAR imaging link; the radar operating parameters enter the SAR imaging link from the parameter acquisition link to obtain the corresponding reconstructed scene. The SAR imaging method based on the mixed norm deep expansion network provided in this embodiment belongs to the SAR imaging link. Specifically, in the SAR imaging link process for the radar operating parameters, the echo matrix can be obtained according to the SAR original echo signals collected at all sampling times, the SAR observation scene can be determined, and the corresponding slant range matrix of each scene point in the observation scene at each random sampling time can be constructed, and the echo signal after the range migration correction is determined according to the echo matrix and the slant range matrix, and the echo signal after the range migration correction is input into the trained mixed norm deep expansion network to obtain the reconstructed scene.

[0131] Based on the same inventive concept, the embodiment of the present application also provides a SAR imaging device based on a mixed norm deep expansion network for implementing the SAR imaging method based on a mixed norm deep expansion network involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the SAR imaging device based on a mixed norm deep expansion network provided below can refer to the limitations of the SAR imaging method based on a mixed norm deep expansion network above, and will not be repeated here.

[0132] In an exemplary embodiment, Figure 3 As shown, a SAR imaging device based on a mixed norm deep expansion network is provided, comprising:

[0133] The radar working parameter acquisition module T1 is used to: acquire radar working parameters; the radar working parameters include the scene center reference slant distance, satellite flight speed, azimuth antenna size, number of transmitted pulses, modulation frequency of the transmitted signal, time vector of the radar transmitting pulse and working wavelength;

[0134] The echo matrix acquisition module T2 is used to: acquire the echo matrix according to the SAR original echo signals collected at all sampling moments;

[0135] The slant range matrix construction module T3 is used to: determine the observation scene of the SAR and construct the corresponding slant range matrix of each scene point in the observation scene at each random sampling moment;

[0136] The echo signal determination module T4 after range migration correction is used to: determine the echo signal after range migration correction according to the echo matrix and the slant range matrix;

[0137] The scene reconstruction module T5 is used to: input the echo signal after the range migration correction into the trained hybrid norm deep expansion network to obtain a reconstructed scene; the hybrid norm deep expansion network includes K levels, each level includes four submodules, namely a first regularization module, a sparse transformation module, a second regularization module and a multiplier update module, and the reconstruction module outputs the result; in any level, the data input in the previous level is used to obtain a reconstructed scene through the reconstruction module, the first regularization module is used to solve the Tikhonov regularization equation to obtain an iterative solution result; the sparse representation of the iterative solution result is obtained through the sparse transformation module; the second regularization module is used to solve L 1 / 2 Regularize the equation and obtain the iterative solution result; the multiplier update module is used to implement the update of the Lagrange multiplier.

[0138] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store SAR imaging processing data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a SAR imaging method based on a mixed norm deep expansion network is implemented.

[0139] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0140] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0141] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0142] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0143] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0144] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0145] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the 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.

[0146] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A SAR imaging method based on a mixed norm deep unfolding network, characterized in that: The SAR imaging method based on the mixed norm deep expansion network includes: Acquire radar operating parameters; the radar operating parameters include scene center reference slant distance, satellite flight speed, azimuth antenna size, number of transmitted pulses, modulation frequency of transmitted signals, time vector of radar transmitted pulses and operating wavelength; Acquire the echo matrix according to the SAR original echo signals collected at all sampling moments; Determine the SAR observation scene and construct the corresponding slant range matrix of each scene point in the observation scene at each random sampling moment; Determine the echo signal after range migration correction according to the echo matrix and the slant range matrix; The echo signal after the range migration correction is input into the trained hybrid norm deep expansion network to obtain a reconstructed scene; the hybrid norm deep expansion network includes K levels, each level includes four submodules, namely a first regularization module, a sparse transformation module, a second regularization module and a multiplier update module, and the reconstruction module outputs the result; in any level, the data input in the previous level is used to obtain a reconstructed scene through the reconstruction module, the first regularization module is used to solve the Tikhonov regularization equation to obtain an iterative solution result; the sparse representation of the iterative solution result is obtained through the sparse transformation module; the second regularization module is used to solve L 1 / 2 Regularize the equation and obtain the iterative solution result; the multiplier update module is used to implement the update of the Lagrange multiplier.

2. The SAR imaging method based on the mixed norm deep expansion network according to claim 1 is characterized in that: The iterative solution result of the first regularization module is expressed as follows: Among them, z (k) represents the k-th iteration solution result of the first regularization module, k represents the number of iterations, ρ (k) and λ (k) represents the optimization parameter of the current k-th iteration, σ (k-1) and (k-1) They respectively represent the iterative update results of the observation scene σ and the Lagrange multiplier y of the k-1th iteration of the previous layer.

3. The SAR imaging method based on the mixed norm deep expansion network according to claim 1 is characterized in that: The sparse transformation module consists of three convolutional layers.

4. The SAR imaging method based on the mixed norm deep expansion network according to claim 1 is characterized in that: The iterative solution results of the second regularization module are expressed as follows: in, In the formula, σ (k) It represents the k-th iteration solution result of the second regularization module, where k represents the number of iterations.

5. The SAR imaging method based on the mixed norm deep expansion network according to claim 1 is characterized in that: Determining the echo signal after range migration correction according to the echo matrix and the slant range matrix specifically includes: Determining a scene delay signal matrix according to the echo matrix and the slant range matrix; Using projection technology, the scene delay signal matrix is ​​enhanced to obtain a projected scene delay signal matrix; The projected scene time-delay signal matrix is ​​summed and superimposed to determine the echo signal after range migration correction.

6. The SAR imaging method based on the mixed norm deep expansion network according to claim 5 is characterized in that: Determining a scene delay signal matrix according to the echo matrix and the slant range matrix specifically includes: Performing a range-to-Fourier transform on the echo matrix to obtain an echo signal after the range-to-Fourier transform; Performing distance-compressed matched filtering on the echo signal after the distance Fourier transform to obtain filtered echo information; Performing a range-direction inverse Fourier transform on the filtered echo information to obtain a range-compressed echo signal; According to the slope distance matrix, using the formula t = r (η m ) / 2c, determine the time delay from SAR transmitting signal to receiving reflected signal; where r(η m ) is at each sampling time η m The corresponding slant distance matrix of each scene point in the observation scene, c is the speed of light; Find the echo signals of each scene point corresponding to the time delay from the echo signals after distance compression; Construct a Na×N 2 The first empty matrix, and the echo signals of each scene point corresponding to the delay are placed in Na×N 2 The first empty matrix of the scene delay signal matrix is ​​obtained; where N a is the number of transmitted pulses, and N is the number of scene points along the x-axis or y-axis of the observed scene.

7. The SAR imaging method based on a mixed norm deep expansion network according to claim 1, characterized in that: Before inputting the range migration corrected echo signal into the trained hybrid norm deep expansion network, the SAR imaging method based on the hybrid norm deep expansion network further includes: Acquire a training data set; the training data set includes a plurality of observation scenes and an echo corresponding to each observation scene; The training data set is used to train the mixed norm deep expansion network to obtain a trained mixed norm deep expansion network.

8. A SAR imaging device based on a mixed norm deep expansion network, characterized in that: The SAR imaging device based on the mixed norm deep expansion network includes: The radar working parameter acquisition module is used to: acquire radar working parameters; the radar working parameters include scene center reference slant distance, satellite flight speed, azimuth antenna size, number of transmitted pulses, modulation frequency of transmitted signal, time vector of radar transmitting pulse and working wavelength; The echo matrix acquisition module is used to: acquire the echo matrix according to the SAR original echo signals collected at all sampling moments; The slant range matrix construction module is used to: determine the SAR observation scene and construct the corresponding slant range matrix of each scene point in the observation scene at each random sampling moment; A module for determining an echo signal after range migration correction, used to: determine an echo signal after range migration correction according to the echo matrix and the slant range matrix; The scene reconstruction module is used to: input the echo signal after the range migration correction into the trained hybrid norm deep expansion network to obtain the reconstructed scene; the hybrid norm deep expansion network includes K levels, each level includes four submodules, namely the first regularization module, the sparse transformation module, the second regularization module and the multiplier update module, and the reconstruction module outputs the result; in any level, the data input in the previous level is used to obtain the reconstructed scene through the reconstruction module, the first regularization module is used to solve the Tikhonov regularization equation to obtain the iterative solution result; the sparse representation of the iterative solution result is obtained through the sparse transformation module; the second regularization module is used to solve L 1 / 2 Regularize the equation and obtain the iterative solution result; the multiplier update module is used to implement the update of the Lagrange multiplier.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the SAR imaging method based on a mixed norm deep unfolding network according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the SAR imaging method based on a mixed norm deep expansion network described in any one of claims 1 to 7 is implemented.

Images