SAR interference intelligent reconstruction method and system based on complex signal domain coding and decoding network

Abstract

The invention discloses an SAR interference intelligent reconstruction method and system based on a complex signal domain coding and decoding network, and the method comprises the steps: S1, constructing an SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network, wherein the model comprises an input preprocessing module, a coding feature extraction module, an improved connection module, a decoding feature fusion module and an image restoration module; s2, constructing a data set for training a model; s3, training the SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network by adopting the training set, and storing a trained weight parameter; s4, pre-stored weight parameters are loaded, and a trained SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network is obtained; and S5, inputting an SAR echo signal to the trained SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network for interference reconstruction. According to the invention, extraction, reconstruction and cancellation of interference signals in SAR signals are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of computer neural network and radar signal processing, and more specifically, to a SAR interference intelligent reconstruction method and system based on a complex signal domain encoding and decoding network. Background Art

[0002] Synthetic Aperture Radar (SAR) is a widely used microwave remote sensing device that actively radiates electromagnetic waves to perform high-resolution imaging of the area to be measured. As the electromagnetic environment becomes more complex, SAR is susceptible to various coherent forwarding interferences during operation, which mask the real target signals in the SAR image and seriously reduce the quality of the SAR image. Interference suppression technology, as a defense method for radar, reduces the impact of interference signals on imaging through corresponding methods, providing support for subsequent image interpretation and other work.

[0003] In the field of optical image processing, deep neural networks can adaptively extract deep features from training set data to complete classification or regression tasks. The radar image formed by SAR is obtained by actively emitting microwaves and receiving their reflected signals, and then compressing and phase correcting them. The image contains rich electromagnetic scattering information. Therefore, SAR images are a unique type of remote sensing data. Their imaging principles, data characteristics, and information presentation methods are very different from those of optical images, resulting in the inability of neural network models or methods that perform well in optical image processing to adapt to SAR image processing. Therefore, when the SAR system is interfered by coherent forwarding, directly performing neural network processing on the radar image does not conform to the relevant principles of SAR signal processing, and processing should be considered on the echo before compression.

[0004] In summary, it is necessary to design a corresponding neural network model or optimize its structure according to the echo characteristics of SAR to achieve the reconstruction of interference echoes and lay the foundation for further interference suppression. Summary of the invention

[0005] The purpose of the present invention is to provide a SAR interference intelligent reconstruction method and system based on a complex signal domain coding and decoding network to overcome the defects of the prior art.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] The SAR interference intelligent reconstruction method based on complex signal domain encoding and decoding network includes the following steps:

[0008] S1. Construct a SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network, which includes an input preprocessing module, a coding feature extraction module, an improved connection module, a decoding feature fusion module and an image restoration module;

[0009] S2. constructing a data set for training the SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network;

[0010] S3, using the training set to train the coding feature extraction module, the improved connection module, and the decoding feature fusion module of the SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network, and saving the weight parameters of the trained coding feature extraction module, the improved connection module, and the decoding feature fusion module;

[0011] S4, loading the pre-stored weight parameters to obtain the trained SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network;

[0012] S5. Input the SAR echo signal to the trained SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network to perform interference reconstruction.

[0013] Furthermore, in step S1:

[0014] An input preprocessing module is used to obtain an input data set, divide the data set into a training set and a test set, randomly scatter and slice the training set and the test set, extract amplitude and phase information from the randomly scatter and sliced ​​training set and the test set, and perform normalization and size adjustment;

[0015] The encoding feature extraction module is used to downsample the input image using convolution and pooling operations to extract the feature matrix;

[0016] An improved connection module, used for performing two convolution operations on the feature matrix using a convolution kernel and then connecting the Dropout layer;

[0017] A decoding feature fusion module, used for upsampling the extracted feature matrix to the input resolution to perform signal sampling point level decomposition, and performing improved connection fusion of high-resolution features of different scales from the encoder;

[0018] The image restoration module is used to re-compress the echo information and calculate the corresponding parameters to obtain the final result.

[0019] Furthermore, the step S3 specifically includes:

[0020] Step 1: Use the Relu function as the activation function in training. The Relu function is as follows, where x is the input sample data value:

[0021]

[0022] Step 2: Forward propagate the sample data to obtain the predicted value, calculate the model loss, and use the Huber function as the loss function to calculate the loss. The formula is:

[0023]

[0024] In the formula, y and f(x) are the true value and predicted value of the sample respectively, and δ is a hyperparameter;

[0025] Step 3: Use the optimizer to update the weight parameters of the model;

[0026] Step 4: Perform training iterations according to the training process from step 1 to step 3, and store the final weight parameters of each module.

[0027] The present invention also provides a system for implementing the above-mentioned SAR interference intelligent reconstruction method based on complex signal domain coding and decoding network, comprising:

[0028] The first building module is used to build a SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network, and the model includes an input preprocessing module, a coding feature extraction module, an improved connection module, a decoding feature fusion module and an image restoration module;

[0029] A second construction module is used to construct a data set for training the SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network;

[0030] A training module, used to train the coding feature extraction module, the improved connection module, and the decoding feature fusion module of the SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network using the training set, and save the weight parameters of the trained coding feature extraction module, the improved connection module, and the decoding feature fusion module;

[0031] A loading module is used to load the pre-stored weight parameters to obtain a trained SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network;

[0032] The input module is used to input the SAR echo signal to the trained SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network for interference reconstruction.

[0033] Compared with the prior art, the advantages of the present invention are as follows: the present invention realizes interference reconstruction by inputting uncompressed SAR complex echo signals, using a codec network in combination with low-resolution information after multiple downsampling and high-resolution information directly transmitted from the coding layer to the decoding layer at the same height through an improved splicing operation, and finally uses echo signals to cancel the reconstructed interference signals and recompress the images, thereby realizing the extraction, reconstruction and cancellation of interference signals in SAR signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is the overall flow chart of the present invention;

[0036] Figure 2 A logical structure diagram of the system constructed in step S1 of the present invention;

[0037] Figure 3 Part of the images of the MiniSAR dataset used in this invention (only the processed SAR grayscale image is shown);

[0038] Figure 4 Part of the FuCheng dataset used in this invention (only the processed SAR grayscale image is shown)

[0039] Figure 5 Part of the images of the COSMO dataset used in this invention (only the processed SAR grayscale image is shown);

[0040] Figure 6 This is a typical sample of the SAR echo signal after fixed frequency shift modulation forwarding used for training the system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0041] Figure 7 This is a typical sample of the SAR echo signal after the intermittent sampling modulation forwarding for the training system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0042] Figure 8 This is a typical sample of the SAR echo signal after the azimuth intermittent sampling modulation forwarding used for the training system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0043] Fig. 9This is a typical sample of the SAR echo signal after two-dimensional intermittent sampling modulation forwarding used for training the system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0044] Fig.10 This is a typical sample of the SAR echo signal after the frequency micro-modulation forwarding used for training the system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0045] Fig.11 This is a typical sample of the SAR echo signal after intermittent sampling + frequency micro-modulation forwarding for the training system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0046] Fig.12 This is a typical sample of the SAR echo signal after intermittent sampling + random frequency shift modulation forwarding used for training the system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0047] Fig.13 This is a typical sample of the SAR echo signal after forwarding by azimuth intermittent sampling + stepped frequency shift modulation for training the system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0048] Fig.14 This is a typical sample of the SAR echo signal after forwarding the micro-motion frequency + step-shift frequency modulation for training the system in the second step of the present invention (only the processed SAR grayscale image is shown);

[0049] Fig.15 This is the structure diagram of the encoding and decoding network;

[0050] Fig.16 Typical test results of the complex signal domain encoding and decoding neural network proposed by the present invention;

[0051] Fig.17 Typical test results for interference reconstruction and image restoration; Fig.18 It is a system framework diagram of the present invention. DETAILED DESCRIPTION

[0052] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0053] See also Figure 1 As shown, this embodiment discloses a SAR interference intelligent reconstruction method based on a complex signal domain encoding and decoding network, comprising the following steps:

[0054] Step S1, constructing a SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network, the model includes an input preprocessing module, a coding feature extraction module, an improved connection module, a decoding feature fusion module and an image restoration module, such as Figure 2 shown.

[0055] In this embodiment, the input preprocessing module is used to obtain an input data set, randomly scatter the samples to enhance the generalization ability and robustness of the model, and divide the data set into a training set and a test set. At the same time, the module enhances the generalization ability and robustness of the model by randomly slicing the samples. Then, the amplitude and phase information of the samples are extracted, and normalization, size adjustment and other operations are performed to facilitate network processing.

[0056] The encoding feature extraction module is also called a compression path or a downsampling path. After the SAR complex echo data to be processed of size 1928×220×2 is sent to the network, two convolution layers of size 3×3×64 are used, followed by a maximum pooling operation using a convolution kernel of size 2×2. This process is defined as the first stage, and then the following network structure repeats the first stage operation process four times. In the repetitive process, the number of convolution kernels begins to change after each pooling operation, and the number increases exponentially, from the initial 64 to 512. When repeating the first stage for the fourth time, a Dropout layer is inserted between the convolution layer and the pooling layer. This process is the encoding stage of the network. In general, the input image is downsampled through a series of convolution and pooling operations to extract the feature matrix.

[0057] The improved connection module is located between the encoder and the decoder, and uses a convolution kernel of size 3×3×1024 to perform two convolution operations, followed by a Dropout layer. Since the feature matrices of the paths on both sides of the codec are of different scales, the improvement is mainly reflected in the normalization of the feature matrix of the coding path by clipping it to the feature matrix of the same size as the decoding path, that is, for the adjustment of the size of the SAR complex echo signal, the specific method is to add a resize layer in front of each node of the decoding module. The function of this module is to further extract and fuse features to provide more useful information for the decoder, especially for the problem that deep neural networks may "forget" certain features when passing information through continuous layers. The improved connection can reintroduce these features to make learning more powerful. This idea was introduced in the residual network (ResNet), and it also shows the advantages of improved classification and smoother learning gradients. Inspired by this mechanism, the improved connection is added to the codec network of the present invention so that each decoder contains the feature matrix of its corresponding encoder.

[0058] The decoding feature fusion module is also called the expansion path or upsampling path, which is symmetrical with the encoder part in data size. The extracted global deep features are upsampled to the input resolution for signal sampling point level decomposition, and the high-resolution features of different scales from the encoder are fused by improved connection to alleviate the loss of spatial information caused by downsampling. After each upsampling step, the feature matrix of the corresponding level of the encoder is spliced ​​with the feature matrix of the current level of the decoder through the improved connection. After the above encoding operation is completed, a 120×13×1024 feature matrix is ​​obtained, and then the expansion path operation is started. This part consists of 4 modules. Before each module starts, the size of the feature matrix is ​​doubled by a 2×2×1024 transposed convolution layer, and the number of convolution kernels is halved. Then, the feature matrix of the corresponding layer of the decoding path is merged through the improved connection, and then two 3×3 convolution layers are used. The size of the feature matrix finally obtained returns to 1928×220×2. The present invention defines this process as the second stage. The entire decoding path needs to repeat the second stage 4 times. Similar to the first stage, the number of convolution kernels begins to change after each splicing operation in the repetitive process, and its number decreases exponentially, from the initial 512 to 64. After the second stage is completed, it is necessary to use 2 convolution kernels of size 1*1 for convolution operation, and the entire encoding and decoding network is completed.

[0059] The image restoration module re-compresses the echo information and images it, calculates the corresponding parameters, and obtains the final result.

[0060] Step S2: constructing a data set for training the SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network.

[0061] In this embodiment, in the scenario set, we constructed a coherent forwarding interference signal model and selected 9 modulation forwarding interference modes in the confrontation scenario. Compared with the usual situation, the innovation is that in addition to considering 5 single modulation methods, we also selected 4 composite modulation methods as more difficult identification options. The specific principle of each modulation is shown in Table 1.

[0062] Table 1 Nine types of modulation and forwarding interference echo signal forms

[0063]

[0064] Among them, t r For fast time, t a For slow time, A 0 is the echo amplitude, ω r (t r ) represents the range antenna pattern function, ω a (t a) represents the range antenna pattern function, R(t r ,t a ) represents the distance from the SAR to the target, c represents the speed of light, rect(·) represents the rectangular signal, t ac is the reference time of the slow time, f c is the carrier frequency of the SAR signal, K r is the range modulation frequency of the linear frequency modulation signal transmitted by the SAR. For frequency shift interference, f d 、f sh is the frequency shift amount; for intermittent sampling forwarding interference, g(t) is the intermittent sampling pulse signal, whose sampling time is τ and repetition period is T s '; For micro-modulation interference, A m is the modulation amplitude, f m is the modulation frequency, is the initial phase of the modulation signal.

[0065] According to the above principle of this embodiment, the MiniSAR dataset, FuCheng dataset and COSMO dataset are used as input, and the typical scenarios are as follows: Figure 3-5 As shown. Through computer simulation, the above three data sets without the implementation of coherent modulation forwarding technology generate coherent modulation forwarding signals to form interference data sets composed of coherent modulation forwarding signals. The typical samples are as follows: Figure 6-Figure 14 As shown. In order to enhance the generalization ability of the network, 5 modulation forwarding positions are set in each coherent modulation forwarding technology, 9 interference states and 4 interference-signal ratios are set at each point, and the parameters of the modulation forwarding technology are randomly generated. The interference-signal ratio is defined as the ratio of the modulation power to the original echo signal power, as shown in formula (1).

[0066]

[0067] Where P J Indicates the modulation power, P S Indicates the original echo signal power.

[0068] Step S3, using the training set to train the coding feature extraction module, improved connection module, and decoding feature fusion module of the SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network, and save the weight parameters of the trained coding feature extraction module, improved connection module, and decoding feature fusion module. The network structure is as follows Fig.15 shown.

[0069] In this embodiment, during the training process, this embodiment uses the Relu function as the activation function, and the Relu function is shown in formula (2). Wherein, x is the input sample data value.

[0070]

[0071] Then, the sample data is forward propagated to obtain the predicted value and the model loss is calculated. The loss function is used to measure the inconsistency between the model's predicted value f(x) and the true value y. One of the loss functions is the mean square error (MSE), which refers to the average of the square of the distance between the model's predicted value f(x) and the sample's true value y, as shown in formula (3). The root mean square error RMSE is defined as shown in formula (4).

[0072]

[0073] In the formula, y i and f(x i ) represent the true value and predicted value of the i-th sample respectively, and m is the number of samples. If there are outliers in the sample, MSE will give higher weights to the outliers, but will sacrifice the prediction effect of other normal data points, which will reduce the overall performance of the model. The second loss function is the mean absolute error (MAE), which refers to the average value of the distance between the model prediction value f(x) and the true value y of the sample, as shown in formula (5):

[0074]

[0075] In the formula, y i and f(x i ) represent the true value and predicted value of the i-th sample, respectively, and m is the number of samples. The MAE curve is V-shaped and continuous, but it is not differentiable at yf(x) = 0, and it is difficult for the computer to solve the derivative. Moreover, the gradient of MAE is equal in most cases, which means that even for a small loss value, its gradient is large. This is not conducive to the convergence of the function and the learning of the model.

[0076] Based on this, this embodiment uses the Huber function as the loss function to calculate the loss, as shown in formula (6).

[0077]

[0078] HuberLoss is a combination of MSE and MAE, including a hyperparameter δ. The value of δ determines the emphasis of HuberLoss on MSE and MAE. When |yf(x)|≤δ, it becomes MSE; when |yf(x)|>δ, it becomes similar to MAE. Therefore, HuberLoss has the advantages of both MSE and MAE, reduces the sensitivity to outliers, and realizes the function of being differentiable everywhere.

[0079] Then, the optimizer is used to update the weight parameters of the system. The RMSProp optimizer is used to solve the problem of the Adagrad algorithm by introducing a decay coefficient, so that the historical information can decay exponentially, thereby avoiding the problem of continuous decline in the learning rate. The exponentially weighted moving average of the squared gradient is used to use a different learning rate for each parameter that is adaptively adjusted according to the most recent gradient size of the parameter, so that the adjustment of the learning rate is smoother. The specific formula is shown in formula (7).

[0080]

[0081] Among them, g is the gradient of the weight parameter, θ is the weight parameter, r is the cumulative square gradient, ρ is the decay rate, ⊙ represents the element product, ε = 10 -8 is the global learning rate, avoiding division by 0.

[0082] Finally, the training iteration is performed according to the above training process, and the final weight parameters of each module are stored.

[0083] Step S4: load the pre-stored weight parameters to obtain a trained SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network.

[0084] Step S5, input the SAR echo signal to the trained SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network for interference reconstruction, perform inferential restoration of the interfered image, and then complete the recompression imaging of the echo signal in the image recovery module, and evaluate the imaging quality.

[0085] In this embodiment, the image quality is evaluated by comparing the visual error between the de-interference processed image and the real image. In addition to the above-mentioned MSE, the peak signal-to-noise ratio (PSNR) is also used. It is an engineering term that represents the ratio of the maximum possible signal power to the destructive noise power that affects its representation accuracy, and is usually expressed in logarithmic decibel units. For a grayscale image, assume that the original image and the de-interference image, both of which are m×n in size, are I 0 and I p , the mean square error MSE is defined as shown in formula (8).

[0086]

[0087] PSNR is calculated by MSE and is defined as shown in formula (9).

[0088]

[0089] In the formula, MAX I To represent the maximum value of the image pixel value, the closer the denoised image is to the original image, the smaller the MSE is, and the larger the PSNR is.

[0090] Then the similarity of the two images is measured again by SSIM. SSIM evaluates image quality by comparing three dimensions: brightness, contrast, and structure, and is particularly suitable for the quality evaluation of the interference suppression result image in the present invention. Of the two images used by SSIM, one is the original image x, and the other is the image y after interference removal. The definition of SSIM is shown in formula (10). Where α, β, γ>0.

[0091] SSIM(x,y)=l(x,y) α c(x,y) β s(s,y) γ (10)

[0092] The brightness contrast function is defined as

[0093]

[0094] In formula (11), μ x is the average grayscale, used for brightness measurement estimation, C 1 is the selected constant.

[0095] The contrast contrast function is defined as:

[0096]

[0097] In formula (12), σ x is the standard deviation, defined as For contrast estimation, C 2 is another selected constant.

[0098]

[0099] The structural contrast function is defined as:

[0100]

[0101] In formula (14), C 3 is another selected constant.

[0102] In summary, in order to obtain a simplified form, let α=β=γ=1, C 3 =C 2 / 2, we get

[0103]

[0104] Complete the final result.

[0105] After testing, the average SSIM of the interference reconstructed by the trained codec network and the ideal interference can reach 81.1%, the average RMSE is only 11.32, and the average PSNR can reach 20.45. Compared with not reconstructing and canceling interference, the similarity between the corresponding scene and the interference-free scene can be increased to 85.5%, the average RMSE is only 4.43, and the average PSNR can reach 28.85.

[0106] The present invention has been verified through simulation tests that it has good interference extraction, reconstruction and cancellation effects for coherent modulation forwarding interference. Its typical training curve is as follows Fig.16 As shown in , the final Loss is 0.02. Fig.17 As shown in the figure, the interference repair effect of SAR complex echo using RMSProp optimizer and Huber loss function is very good.

[0107] See also Fig.18 As shown, the present invention also provides a system for implementing the above-mentioned SAR interference intelligent reconstruction method based on the complex signal domain codec network, comprising: a first construction module 1, used to construct a SAR interference intelligent reconstruction model based on the complex signal domain codec network, the model comprising an input preprocessing module, a coding feature extraction module, an improved connection module, a decoding feature fusion module and an image restoration module; a second construction module 2, used to construct a data set for training the SAR interference intelligent reconstruction model based on the complex signal domain codec network; a training module 3, used to use the training set to train the coding feature extraction module, the improved connection module, and the decoding feature fusion module of the SAR interference intelligent reconstruction model based on the complex signal domain codec network, and save the weight parameters of the trained coding feature extraction module, the improved connection module, and the decoding feature fusion module; a loading module 4, used to load the pre-stored weight parameters to obtain the trained SAR interference intelligent reconstruction model based on the complex signal domain codec network; an input module 5, used to input the SAR echo signal to the trained SAR interference intelligent reconstruction model based on the complex signal domain codec network for interference reconstruction.

[0108] Therefore, it can be seen that the complex signal domain encoding and decoding network can complete the extraction, reconstruction and cancellation functions of SAR composite modulation forwarding interference with higher SSIM and PSNR, and can retain the essential characteristics of the SAR signal and reduce the number of overfitting, which verifies that the present invention has a good effect.

[0109] Although the embodiments of the present invention are described in conjunction with the accompanying drawings, the patent owner may make various variations or modifications within the scope of the appended claims. As long as they do not exceed the protection scope described in the claims of the present invention, they should be within the protection scope of the present invention.

Claims

1. A SAR interference intelligent reconstruction method based on a complex signal domain encoding and decoding network, characterized in that: The following steps are involved: S1. Construct a SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network, which includes an input preprocessing module, a coding feature extraction module, an improved connection module, a decoding feature fusion module and an image restoration module; S2. constructing a data set for training the SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network; S3, using the training set to train the coding feature extraction module, the improved connection module, and the decoding feature fusion module of the SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network, and saving the weight parameters of the trained coding feature extraction module, the improved connection module, and the decoding feature fusion module; S4, loading the pre-stored weight parameters to obtain the trained SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network; S5. Input the SAR echo signal to the trained SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network to perform interference reconstruction.

2. The SAR interference intelligent reconstruction method based on complex signal domain coding and decoding network according to claim 1 is characterized in that: In step S1: An input preprocessing module is used to obtain an input data set, divide the data set into a training set and a test set, randomly scatter and slice the training set and the test set, extract amplitude and phase information from the randomly scatter and sliced ​​training set and the test set, and perform normalization and size adjustment; The encoding feature extraction module is used to downsample the input image using convolution and pooling operations to extract the feature matrix; An improved connection module, used for performing two convolution operations on the feature matrix using a convolution kernel and then connecting the Dropout layer; A decoding feature fusion module, used for upsampling the extracted feature matrix to the input resolution to perform signal sampling point level decomposition, and performing improved connection fusion of high-resolution features of different scales from the encoder; The image restoration module is used to re-compress the echo information and calculate the corresponding parameters to obtain the final result.

3. The SAR interference intelligent reconstruction method based on complex signal domain coding and decoding network according to claim 1 is characterized in that: The step S3 specifically includes: Step 1: Use the Relu function as the activation function in training. The Relu function is as follows, where x is the input sample data value: Step 2: Forward propagate the sample data to obtain the predicted value, calculate the model loss, and use the Huber function as the loss function to calculate the loss. The formula is: In the formula, y and f(x) are the true value and predicted value of the sample respectively, and δ is a hyperparameter; Step 3: Use the optimizer to update the weight parameters of the model; Step 4: Perform training iterations according to the training process from step 1 to step 3, and store the final weight parameters of each module.

4. A system for implementing the SAR interference intelligent reconstruction method based on a complex signal domain coding and decoding network according to any one of claims 1 to 3, characterized in that: include: The first building module is used to build a SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network, and the model includes an input preprocessing module, a coding feature extraction module, an improved connection module, a decoding feature fusion module and an image restoration module; A second construction module is used to construct a data set for training the SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network; A training module, used to train the coding feature extraction module, the improved connection module, and the decoding feature fusion module of the SAR interference intelligent reconstruction model based on the complex signal domain coding and decoding network using the training set, and save the weight parameters of the trained coding feature extraction module, the improved connection module, and the decoding feature fusion module; A loading module is used to load the pre-stored weight parameters to obtain a trained SAR interference intelligent reconstruction model based on a complex signal domain encoding and decoding network; The input module is used to input the SAR echo signal to the trained SAR interference intelligent reconstruction model based on the complex signal domain encoding and decoding network for interference reconstruction.

Images