A Secure Distributed SAR Image Data Generation Method Based on Split Federated Learning

Abstract

The present invention discloses a secure distributed SAR image data generation method based on split federated learning, belonging to the technical field of SAR images, specifically including: constructing a split federated learning model including multiple clients; on each client, applying a data segmentation algorithm to split the locally held SAR image dataset into several subsets; each client independently trains a local generation model on its subset and generates pseudo synthetic aperture radar image data; using the federated learning framework, each client only shares the parameter updates of its local generation model with the central server; the central server aggregates the received model parameter updates to generate global model parameters; the central server distributes the global model parameters to all clients, and each client updates its local generation model according to the global model parameters; repeating the above steps until a predetermined number of training iterations is reached; the present invention improves the generalization ability and accuracy of the global model.

Description

Technical Field

[0001] The present invention relates to the technical field of SAR image, and particularly relates to a secure distributed SAR image data generation method based on split federated learning. Background Art

[0002] Synthetic Aperture Radar (SAR) is an advanced remote sensing technology, which is widely used in the fields of earth science research, disaster monitoring, military reconnaissance, etc. Due to its unique imaging mechanism and rich information content, SAR image data plays an irreplaceable role in many application scenarios. However, with the rapid increase in the volume of SAR image data and the diversification of application requirements, how to efficiently and securely generate and process these data has become an important research topic.

[0003] Traditional SAR image data processing methods usually rely on a centralized data processing center, and all data needs to be transmitted to the data center for processing. This method not only places extremely high requirements on network bandwidth and storage capacity, but also has significant security risks during data transmission. Once the data is leaked or tampered with, serious consequences may occur.

[0004] In recent years, federated learning, as an emerging distributed machine learning framework, provides a new idea for solving the above problems. Federated learning allows multiple participants to collaborate in learning by sharing model parameter updates instead of raw data while keeping the data stored locally. This method can not only effectively protect data privacy, but also utilize the data resources of each participant to improve the training effect of the model.

[0005] Although federated learning has many advantages in theory, it still faces some challenges in practical applications. First, the complexity and high-dimensional characteristics of SAR image data make it challenging to directly apply federated learning algorithms in terms of computation and communication. Second, existing federated learning algorithms often assume that the data distributions of all participants are the same, while in fact, SAR image data from different sources may have significant differences, which will affect the generalization ability and training efficiency of the model. Split Federated Learning (SFL) is an emerging distributed learning technology that combines the advantages of federated learning and split learning. It improves the efficiency of model training and communication while protecting data privacy, and is applicable to various scenarios that require distributed data collaborative learning. Therefore, the present invention proposes a secure distributed SAR image data generation method based on split federated learning. Summary of the Invention

[0006] The purpose of the present invention is to provide a secure distributed SAR image data generation method based on split federated learning, and solve the following technical problems:

[0007] SAR image data from different sources may have significant differences, which can affect the generalization ability and training efficiency of the model.

[0008] The objective of the present invention can be achieved by the following technical solutions:

[0009] A secure distributed SAR image data generation method based on split federated learning, comprising the following steps:

[0010] Step 1: Construct a split federated learning model, which includes a central server and multiple clients distributed at different geographical locations. Each client holds a part of the SAR image dataset, and there is no interaction between the image datasets.

[0011] Step 2: On each client, apply a data splitting algorithm to split the locally held SAR image dataset into several subsets.

[0012] Step 3: Based on the local data subsets, each client independently trains a local model using a generative adversarial network, performs feature extraction and parameter adjustment, and uses the local model to process new inputs, thereby generating pseudo-synthetic aperture radar images as output results.

[0013] Step 4: After each client completes one round of local optimization, calculate the change in the model parameters during the current iteration cycle. Using the federated learning framework, each client only shares the parameter updates of its locally generated model with the central server or other clients.

[0014] Step 5: The central server or other clients aggregate the received model parameter updates, verify the parameters, and generate global model parameters based on federated averaging.

[0015] Step 6: The central server evaluates the performance of the newly generated global model parameters on a preset validation set, distributes the qualified global model parameters to all clients, and each client updates its respective local generated model according to the global model parameters. After the client completes local training, calculate the change in the model parameters again and send these updates back to the central server for use in the next round of aggregation.

[0016] Step 7: Repeat steps 3 to 6 until a predetermined number of training iterations is reached.

[0017] As a further solution of the present invention: When transmitting model parameter updates between each client and the central server, use end-to-end encryption technology, introduce differential privacy technology in the process of parameter updates of the locally generated model, and add noise to the data samples.

[0018] As a further solution of the present invention: In step 2, the specific process of the data splitting algorithm is as follows:

[0019] Regard the entire SAR image dataset as an overall region Ω, calculate the average gray value and standard deviation of all pixel points within the region; divide the overall region into several sub-regions, and for each sub-region R i , calculate the average gray value μ i and standard deviation σ i；

[0020] The formula for calculating the average value is: μ = (∑x j ) / N, where x j is the gray value of the j-th pixel point within the sub-region R, and N is the number of pixel points within the sub-region;

[0021] The formula for calculating the standard deviation is: σ = √[(∑(x j -μ)^2) / (N - 1)];

[0022] Set a threshold T. When σ i > T, then perform splitting; otherwise, do not split; split the sub-region R i that meets the splitting condition into four smaller sub-regions R i1 , R i2 , R i3 , R i4 , and continue to determine whether each newly generated sub-region needs to be split until the σ values of all sub-regions are less than or equal to the threshold T;

[0023] For any two sub-regions A and B after splitting, if D(A, B) < S, then re-combine sub-regions A and B, where D represents the distance metric between the two regions, and S is a preset merging threshold.

[0024] As a further solution of the present invention: in the step 3, the process of generating pseudo-synthetic aperture radar image data is as follows:

[0025] Denoise, normalize, and enhance the original SAR image data, extract features from the preprocessed SAR image, the features include texture features and spectral features, construct a generation model based on a generative adversarial network, use the extracted features as input, train the selected generation model, and for each subset D i , calculate the loss function L(θ), where θ represents the model parameters;

[0026]

[0027] where D is the discriminator, G is the generator, p data is the real data distribution, p zis the noise distribution; the model parameters θ are updated using the gradient descent method to minimize the loss function L(θ); new pseudo-SAR image data G(θ, z) is generated, where z represents a random noise vector.

[0028] As a further aspect of the present invention: the training process of the generative adversarial network further includes:

[0029] Fix the discriminator D and train the generator G to maximize the probability of deceiving the discriminator;

[0030] Fix the generator G and train the discriminator D to improve its ability to distinguish between real images and generated images;

[0031] Alternate the above two steps until a Nash equilibrium state is reached.

[0032] As a further aspect of the present invention: in step 4, the process of sharing parameter updates is as follows:

[0033] The parameters are encrypted and transmitted via the TLS protocol. Each client uploads parameter updates at different time points. The central server receives the encrypted parameter updates from each client and stores them in separate memories. Only when the central server receives a sufficient number of parameter updates does it aggregate the parameter updates.

[0034] As a further aspect of the present invention: in step 5, the central server receives the local updates from all clients and calculates the weighted average of these updates. Each client's update is assigned a different weight according to the size of its local data, and the calculated weighted average update is used to update the parameters of the global model.

[0035] As a further aspect of the present invention: in step 6, the process by which each client updates its local generative model according to the global model parameters is as follows:

[0036] Each client receives the latest global model parameters W t , where t represents the current iteration round. The client synchronizes the received global model parameters with its local model parameters. The formula is:

[0037] W local t = αW local t-1 +(1 - α)W t ;

[0038] where α is a weight coefficient between 0 and 1;

[0039] Using the updated local model parameters, the client continues to train the model on its local dataset; assuming the loss function is denoted as l, the training process can be expressed as minimizing the loss function:

[0040] W local t+1 = argmin l(W, D local );

[0041] where D local is the local dataset of the client;

[0042] After completing local training, the client calculates the updated or gradient of the new model parameters. Using the gradient descent method, the update can be expressed as: ΔW local t+1 = -ηabla W l(W local t , Dlocal), where η is the learning rate and abla W l is the gradient of the loss function with respect to the model parameters;

[0043] The client then encrypts and uploads the updated ΔW of the new model parameters local t+1 back to the central server or other specified clients.

[0044] Advantages of the present invention:

[0045] The present invention constructs a federated learning model including multiple participants, and applies a data splitting algorithm on each participant to split the locally held SAR image dataset into several subsets. Locally generated models are independently trained on each subset, and the global model parameters are aggregated and distributed using the federated learning framework. This design not only avoids the direct sharing of raw data and protects data privacy, but also enhances the security of the system by introducing end-to-end encryption technology and differential privacy technology. In addition, this method allows clients to be dynamically added or removed, improving the scalability and flexibility of the system. By independently training the generative model on each subset, it helps to capture local features in the data, thereby improving the generalization ability and accuracy of the global model. At the same time, using the computing resources of multiple clients for distributed training improves the efficiency and speed of data processing, especially suitable for the processing of large-scale SAR image datasets. Overall, the present invention has significant advantages in protecting data privacy, improving system security, improving processing efficiency and flexibility, and provides an effective solution for distributed SAR image data generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present invention will be further described below with reference to the accompanying drawings.

[0047] Figure 1 It is a schematic diagram of the process of the present invention. Specific implementation manners

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the protection scope of the present invention.

[0049] Please refer to Figure 1 As shown, the present invention is a secure distributed SAR image data generation method based on split federated learning. The method aims to improve the security and privacy protection capabilities of the model by using multiple geographically dispersed data sources, while ensuring the quality and diversity of the generated SAR images. The specific implementation process includes the following steps:

[0050] Step 1: Construct a split federated learning model

[0051] Architecture design: Establish a system structure consisting of a central server and multiple clients located in different geographical locations. Among them, each client has its own unique part of the SAR image dataset.

[0052] Data isolation: Ensure that the datasets between clients are independent of each other and do not perform any form of data exchange or sharing to enhance the privacy of the overall system.

[0053] Communication protocol: Develop a secure and efficient communication protocol to ensure the security and reliability of data transmission during the federated learning process. This includes using encryption technology to protect data from being stolen or tampered with during transmission.

[0054] Step 2: Split the local dataset

[0055] Algorithm application: Inside each client, use specific data splitting techniques to divide the original SAR image set it owns into several smaller subsets. This not only helps to accelerate the subsequent processing process but also better adapts to the computational resource requirements of different scales.

[0056] Data preprocessing: Perform necessary preprocessing operations on the divided data subsets, such as normalization, denoising, etc., to improve the data quality and thus enhance the training effect of the model.

[0057] Step 3: Local model training

[0058] Feature extraction and parameter adjustment: For each partitioned dataset, use the generative adversarial network (GAN) as the basic framework to conduct in-depth analysis and learning locally on the client side, extract key features from it, and adjust the model parameter settings accordingly.

[0059] Pseudo-image generation: Use the optimized local model to process new input samples, so as to efficiently produce high-quality synthetic aperture radar images.

[0060] Hyperparameter tuning: During the local training process, continuously adjust the hyperparameters of the generative adversarial network, such as the learning rate, batch size, etc., to achieve the best training effect.

[0061] Step 4: Parameter update and sharing

[0062] Calculation of change amount: After completing one round of iteration, each client needs to count the change situation of its model weights in this round.

[0063] Federated synchronization: Through a secure channel, only send the calculated change information rather than the actual data itself to other participants or directly upload it to the central node. This method effectively avoids the risk of sensitive information leakage.

[0064] Differential privacy: Introduce differential privacy technology during the parameter update process to further protect the privacy of user data and prevent potential privacy leakage problems.

[0065] Step 5: Global model integration

[0066] Parameter aggregation: After receiving information from all participating parties, the central node is responsible for summarizing this information and determining the new round of global optimal solution according to specific rules (such as weighted average).

[0067] Performance verification: In addition, the solution will also be comprehensively tested on a pre-prepared test set to confirm whether it meets the established standard requirements.

[0068] Anomaly detection: During the parameter aggregation process, add an anomaly detection mechanism to identify and eliminate possible anomalies or malicious updates to ensure the stability and security of the global model.

[0069] Step 6: Model parameter distribution and retraining

[0070] Result feedback: Once it is confirmed that the current version has achieved the expected effect, it will be distributed to all relevant parties.

[0071] Continuous improvement: After receiving the latest guidelines, each site will continue to carry out the next round of work according to a similar process until the preset maximum number of iterations is reached.

[0072] Version management: Introduce a version management mechanism to record the updates of model parameters in each round for rollback or comparative analysis when needed.

[0073] Step 7: Iterate until convergence

[0074] Repeated execution: The entire project will continuously repeat the operations in Steps 3 to 6 until the predetermined number of training rounds is reached or it is observed that the performance no longer improves significantly.

[0075] Dynamic adjustment: Dynamically adjust the training strategy according to the actual training situation, such as changing the learning rate, adjusting the batch size, etc., to accelerate the convergence process.

[0076] Monitoring and logging: Establish a complete monitoring system to track the training progress and performance metrics in real time and record detailed log information for later analysis and troubleshooting.

[0077] Through the above series of steps, this solution can fully utilize the advantages of cross-regional cooperation on the premise of ensuring data security, and promote the development of more accurate and reliable SAR image generation technology.

[0078] In a preferred embodiment of the present invention, in order to further enhance the security and privacy protection capabilities of the system, we adopt end-to-end encryption technology when transmitting model parameter updates between each client and the central server. This means that every communication from the client to the central server will be encrypted, ensuring that even if the data is intercepted during transmission, it cannot be interpreted or tampered with by unauthorized third parties. This encryption mechanism provides a secure data transmission environment for the entire federated learning process and greatly reduces the risk of data leakage.

[0079] In addition, in order to enhance the privacy of local data processing, we introduce differential privacy technology in the process of generating model parameter updates locally. Differential privacy is a powerful privacy protection method that protects the privacy information of users from being leaked by making small, randomized modifications (usually called adding noise) to the data to mask the contribution of individual data. In our method, whenever a client updates its local model parameters, a certain amount of noise will be injected into these parameters, so that even if an attacker can access some model parameters, it is very difficult to infer specific data sample information.

[0080] At the same time, we also directly add noise to the data samples as another level of privacy protection measure. This method can further confuse the data features and make it more difficult for any attempt to reverse engineer the original data by analyzing the model output. In this way, we not only protect the information of individual data points but also enhance the privacy protection level of the entire data set.

[0081] In another preferred embodiment of the present invention, in step 2, the specific process of the data segmentation algorithm is as follows:

[0082] Regard the entire SAR image dataset as an overall region Ω, calculate the average gray value and standard deviation of all pixel points within the region; divide the overall region into several sub-regions, and for each sub-region R i , calculate the average gray value μ i and standard deviation σ i；

[0083] The formula for calculating the average value is: μ = (∑x j ) / N, where x j is the gray value of the j-th pixel point within the sub-region R, and N is the number of pixel points within the sub-region;

[0084] The formula for calculating the standard deviation is: σ = √[(∑(x j -μ)^2) / (N - 1)];

[0085] Set a threshold T. When σ i > T, then perform splitting; otherwise, do not split; split the sub-region R i that meets the splitting condition into four smaller sub-regions R i1 , R i2 , R i3 , R i4 , and continue to determine whether each newly generated sub-region needs to be split until the σ values of all sub-regions are less than or equal to the threshold T;

[0086] For any two sub-regions A and B after splitting, if D(A, B) < S, then re-combine sub-regions A and B, where D represents the distance metric between the two regions, and S is a preset merging threshold.

[0087] In another preferred embodiment of the present invention, it is characterized in that in step 3, the process of generating pseudo-synthetic aperture radar image data is as follows:

[0088] Denoise, normalize, and enhance the original SAR image data, extract features from the preprocessed SAR image, where the features include texture features and spectral features, construct a generation model based on a generative adversarial network, use the extracted features as input, train the selected generation model, and for each subset D i , calculate the loss function L(θ), where θ represents the model parameters;

[0089]

[0090] where D is the discriminator, G is the generator, p data is the true data distribution, pz is the noise distribution; the model parameter θ is updated using the gradient descent method to minimize the loss function L(θ); new pseudo-SAR image data G(θ,z) is generated, where z represents a random noise vector.

[0091] In a preferred case of this embodiment, the training process of the generative adversarial network further includes:

[0092] First, we enter the first stage, that is, fixing the discriminator D and focusing on training the generator G. In this stage, the task of the generator G is to generate SAR image data that is as realistic as possible, and its goal is to maximize the probability of deceiving the discriminator D. In other words, the generator G tries to produce pseudo-images that look indistinguishable from real SAR images, so that the discriminator D cannot tell which are real and which are created by the generator G. To achieve this, the generator G continuously adjusts its internal parameters and minimizes a specific loss function through the backpropagation algorithm, which measures the difference between the generated images and the real image distribution.

[0093] Immediately afterwards, the training process enters the second stage. At this time, it is the turn to fix the generator G and instead train the discriminator D. In this stage, the role of the discriminator D changes to a strict "quality inspector", and its task is to improve its ability to distinguish between real SAR images and pseudo-images generated by the generator G. This means that the discriminator D needs to become more sensitive and be able to capture even the slightest differences to correctly classify the source of the input images. Similarly, the discriminator D also optimizes its own parameters through backpropagation, but its goal is to maximize a different loss function, which reflects the accuracy of the discriminator D in classifying true and false images.

[0094] These two stages are not carried out in isolation, but are executed alternately, forming a closed-loop optimization process. In each iteration, the generator G and the discriminator D are adjusted according to each other's current state. This mutually promoting competition mechanism drives both to continuously evolve until a state called Nash equilibrium is reached. At the Nash equilibrium point, the quality of the images generated by the generator G is so high that the discriminator D can no longer effectively distinguish between true and false, and at the same time the discriminator D has become so powerful that it is difficult for the generator G to further improve its deception ability. At this time, we can consider that the GAN training has reached a relatively stable and efficient working state and can produce high-quality SAR image data.

[0095] In another preferred embodiment of the present invention, in step 4, the process of sharing parameter updates is as follows:

[0096] The parameters are encrypted and transmitted via the TLS protocol. Each client uploads parameter updates at different time points. The central server receives the encrypted parameter updates from each client and stores them in an independent memory. Only after the central server receives a sufficient number of parameter updates does it aggregate the parameter updates.

[0097] In another preferred embodiment of the present invention, in step 5, the central server receives the local updates from all clients and calculates the weighted average of these updates. The update of each client is assigned a different weight according to the size of its local data, and the calculated weighted average update is used to update the parameters of the global model.

[0098] In another preferred embodiment of the present invention, in step 6, the process by which the client updates its local generation model according to the global model parameters is as follows:

[0099] Each client receives the latest global model parameters W t , where t represents the current iteration round. The client synchronizes the received global model parameters with the local model parameters. The formula is:

[0100] W local t =αW local t-1 +(1 - α)W t ;

[0101] where α is a weight coefficient between 0 and 1;

[0102] Using the updated local model parameters, the client continues to train the model on its local dataset. Assuming the loss function is labeled as l, the training process can be expressed as minimizing the loss function:

[0103] W local t+1 =argminl(W,D local );

[0104] where D local is the local dataset of the client;

[0105] After completing the local training, the client calculates the new model parameter updates or gradients. Using the gradient descent method, the update can be expressed as: ΔW local t+1 =-ηabla W l(W local t ,Dlocal), where η is the learning rate, and abla W l is the gradient of the loss function with respect to the model parameters;

[0106] The client updates the new model parameter ΔW again local t+1 After encryption, it is uploaded back to the central server or other specified clients.

[0107] The above has described an embodiment of the present invention in detail, but the described content is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A secure distributed SAR image data generation method based on split federated learning, characterized in that, It includes the following steps: Step 1: Construct a split federated learning model, which includes a central server and multiple clients distributed at different geographical locations. Each client holds a part of the SAR image dataset, and there is no interaction between the image datasets; Step 2: On each client, apply a data segmentation algorithm to segment the locally held SAR image dataset into several subsets; Step 3: Based on the local data subsets, each client independently trains a local model using a generative adversarial network, performs feature extraction and parameter adjustment, and uses the local model to process new inputs, thereby generating pseudo-synthetic aperture radar images as output results; Step 4: After each client completes a round of local optimization, calculate the change in the model parameters during the current iteration cycle. Using the federated learning framework, each client only shares the parameter updates of its locally generated model with the central server or other clients; Step 5: The central server or other clients aggregate the received model parameter updates, verify the parameters, and generate global model parameters based on federated averaging; Step 6: The central server evaluates the performance of the newly generated global model parameters on a preset validation set, distributes the qualified global model parameters to all clients. Each client updates its local generated model according to the global model parameters. After the client completes local training, calculate the change in the model parameters again and send these updates back to the central server for use in the next round of aggregation; Step 7: Repeat steps 3 to 6 until a predetermined number of training iterations is reached; In step 2, the specific process of the data segmentation algorithm is: Regard the entire SAR image dataset as a whole region Ω, and calculate the average and standard deviation of the gray values of all pixel points within the region; divide the whole region into several sub-regions, and for each sub-region R i , calculate the average gray value μ i of its internal pixel points and the standard deviation σ i ; The formula for calculating the average value is: μ = (∑x j ) / N, where x j is the grayscale value of the j-th pixel point in the sub-region R, and N is the number of pixel points in the sub-region; The standard deviation calculation formula is: σ = √[(∑(x j - μ)^2) / (N - 1)]; Set a threshold T. When σ i > T, perform splitting; otherwise, do not split. Split the sub-region R i that meets the splitting condition into four smaller sub-regions R i1 , R i2 , R i3 , R i4 . Continue to determine whether each newly generated sub-region needs to be split until the σ values of all sub-regions are less than or equal to the threshold T; For any two sub-regions A and B after splitting, if D(A, B) < S, then re-combine sub-regions A and B, where D represents the distance metric between the two regions, and S is a preset merging threshold; In step 6, the process by which the client updates its local generated model according to the global model parameters is: Each client receives the latest global model parameters \(W\) from the central server t , where \(t\) represents the current iteration round. The client synchronizes the received global model parameters with the local model parameters, and the formula is: ； where α is a weight coefficient between 0 and 1; Using the updated local model parameters, the client continues to train the model on its local dataset; assuming the loss function is labeled as l , the training process is represented as minimizing the loss function: ； Among them, D local is the local data set of the client; After completing local training, the client calculates the updated new model parameters or gradients. Using the gradient descent method, the update is expressed as: ΔW local t+1 =-ηabla W l (W local t ,Dlocal), where η is the learning rate, and abla W l is the gradient of the loss function with respect to the model parameters; The client updates the new model parameter ΔW again local t+1 After encryption, it is uploaded back to the central server or other specified clients.

2. The secure distributed SAR image data generation method based on split federated learning according to claim 1, wherein, When transmitting model parameter updates between each client and the central server, end-to-end encryption technology is used, differential privacy technology is introduced during the parameter update process of the locally generated model, and noise is added to the data samples.

3. A secure distributed SAR image data generation method based on split federated learning according to claim 1, characterized in that, In step 3, the process of generating pseudo-synthetic aperture radar image data is: Denoise, normalize, and enhance the original SAR image data, extract features from the preprocessed SAR image, where the features include texture features and spectral features, construct a generative model based on a generative adversarial network, use the extracted features as input, and train the selected generative model. For each subset D i , calculate the loss function L(θ), where θ represents the model parameters; ； where D is the discriminator, G is the generator, p data is the real data distribution, p z is the noise distribution; Use the gradient descent method to update the model parameter θ to minimize the loss function L(θ); generate new pseudo-SAR image data G(θ, z), where z represents a random noise vector.

4. A secure distributed SAR image data generation method based on split federated learning according to claim 3, characterized in that, The training process of the generative adversarial network also includes: Fix the discriminator D and train the generator G to maximize the probability of deceiving the discriminator; Fix the generator G and train the discriminator D to improve its ability to distinguish between real images and generated images; Alternate between the above two steps until a Nash equilibrium state is reached.

5. A secure distributed SAR image data generation method based on split federated learning according to claim 1, characterized in that, In step 4, the process of sharing parameter updates is: The parameters are encrypted and transmitted via the TLS protocol. Each client uploads parameter updates at different time points. The central server receives the encrypted parameter updates from each client and stores them in an independent memory. Only when the central server has received a sufficient number of parameter updates does it aggregate the parameter updates.

6. The method for generating secure distributed SAR image data based on split federated learning according to claim 1, wherein In step 5, the central server receives the local updates from all clients and calculates the weighted average of these updates. The update of each client is assigned a different weight according to the size of its local data, and the calculated weighted average update is used to update the parameters of the global model.

Images