Multi-client collaborative sar image target recognition method based on federated learning
By using a federated learning approach to transfer model parameters among multiple satellite agencies instead of raw data, the problems of insufficient data sample quantity and data privacy protection in SAR image target recognition are solved. This enables collaborative training of models without sharing data, thereby improving the model's generalization ability and recognition performance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- FUDAN UNIVERSITY
- Filing Date
- 2024-11-22
- Publication Date
- 2026-07-14
Smart Images
Figure CN119516406B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of remote sensing synthetic aperture radar applications, and in particular to a multi-client cooperative SAR image target recognition method based on federated learning. Background Technology
[0002] Synthetic Aperture Radar (SAR) can perform radar imaging under various weather conditions (such as cloud cover, rainfall, etc.) and both day and night. With the advent of the big data era, deep learning has become a major method for data analysis, demonstrating excellent application results in fields such as Earth observation and remote sensing.
[0003] Deep learning-based SAR target recognition methods utilize deep learning to extract features and classify SAR images, achieving high-precision target identification. Compared to traditional methods, deep learning-based methods offer the following advantages: 1. Automatic feature extraction: Deep learning models can automatically extract effective features from data, significantly reducing reliance on manual intervention. 2. Strong generalization ability: Through large-scale data training, deep learning models can learn features with good generalization ability, maintaining high recognition performance in different scenarios and environments. 3. Ability to handle complex scenes: Deep learning models can handle complex backgrounds and multi-target scenes, exhibiting stronger robustness and adaptability. Therefore, deep learning-based SAR target recognition and classification methods have significant application value in multiple fields such as military reconnaissance, disaster monitoring, and environmental monitoring.
[0004] However, although deep learning models have demonstrated powerful recognition capabilities in SAR target identification, these models typically rely on large amounts of data for training. In intelligent interpretation of SAR images, the high cost and difficulty of data acquisition often result in insufficient data samples for single satellite organizations, severely limiting the improvement of model performance and the expansion of its application scope.
[0005] To improve model performance, data sharing through collaboration among multiple satellite agencies has become an effective approach. Increasing the quantity and diversity of samples significantly enhances the model's generalization ability. However, SAR image data contains sensitive geographic location and target information, involving national security and commercial secrets. Improper handling could lead to serious consequences, making data privacy protection crucial during data sharing. Each client (e.g., different satellite agencies) must ensure the confidentiality and security of data during sharing to prevent the leakage of sensitive information. Therefore, data privacy protection becomes a key challenge in multi-client collaborative SAR target recognition systems. Summary of the Invention
[0006] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a multi-client collaborative SAR image target recognition method based on federated learning. By passing model parameters between participants instead of raw data, multiple satellite agencies can collaboratively train the model without sharing data. This effectively ensures model performance while protecting data privacy, and can gradually approach the performance of centralized training through iterative optimization.
[0007] The objective of this invention can be achieved through the following technical solutions:
[0008] A multi-client cooperative SAR image target recognition method based on federated learning includes the following steps:
[0009] S1: Acquire SAR image data and preprocess it to generate corresponding target recognition labels, thus obtaining a labeled dataset;
[0010] S2: Build a federated learning framework, which includes a server and multiple clients participating in collaborative training. Each client shares model parameters with the server; construct a neural network model for object recognition, initialize it, and distribute global model parameters to each client.
[0011] S3: Allocate local private data to each client based on the labeled dataset, and each client uses the local private data to train the local model and uploads the updated model parameters to the server;
[0012] S4: The server aggregates the model parameters uploaded by all clients, updates the global model parameters, and distributes the updated global model parameters to each client;
[0013] S5: Repeat steps S3 and S4 until the global model parameters reach convergence and the trained global model is obtained.
[0014] S6: Test the trained global model and select the global model that passes the test for SAR image target recognition.
[0015] Furthermore, step S2 specifically includes the following steps:
[0016] S201: Establish a semi-decentralized federated learning framework, determine the participating clients and servers, as well as the number of clients, and each client shares model parameters with the server, which aggregates the model parameters of each client to update the global model;
[0017] S202: Pre-build a neural network model for target recognition and initialize global model parameters;
[0018] S203: The server distributes the initialized global model parameters to each client.
[0019] Furthermore, step S3 specifically includes the following steps:
[0020] S301: Each client loads the global model parameters transmitted by the server;
[0021] S302: Each client holds a local dataset allocated based on the labeled dataset. Each client performs local training based on the allocated global model parameters, calculates the loss function, and updates the gradient. The corresponding update expression for the local model parameters is:
[0022]
[0023] In the formula, These are the local model parameters at the (t+1)th iteration. Here are the local model parameters at the t-th iteration, and η is the learning rate. For model parameters The loss function calculation results are shown below, where l is the cross-entropy loss function and D is the loss function. i For the i-th local dataset, (x i,j ,y i,j ( ) represents local data and corresponding target identification labels;
[0024] S303: Each client sends the updated local model parameters to the server.
[0025] Furthermore, step S4 specifically includes the following steps:
[0026] S401: The server receives local model parameters transmitted from each client;
[0027] S402: The server performs aggregation operations on the local model parameters of each client according to a predetermined aggregation algorithm to generate new global model parameters;
[0028] S403: The server updates the global model parameters based on the new global model parameters;
[0029] S404: The server distributes the updated global model parameters to each client.
[0030] Furthermore, the aggregation algorithm is the FedAvg algorithm, which is used to perform a weighted average operation on the model parameters of all clients. The calculation expression of the FedAvg algorithm is:
[0031]
[0032] In the formula, Let N be the global model parameters for the (t+1)th iteration, and N be the number of clients, |Di | represents the data volume of the i-th client. The total amount of data, These are the local model parameters for the i-th client during the t-th iteration.
[0033] Furthermore, step S5 specifically includes the following steps:
[0034] S501: The server distributes the updated global model parameters to each client;
[0035] S502: Each client executes step S3 to train the local model;
[0036] S503: The server executes step S4 to aggregate and update the global model;
[0037] S504: Iteratively execute steps S501-S503 until the global model parameters converge.
[0038] Furthermore, step S1 specifically includes the following steps:
[0039] S101: Acquire SAR image data via satellite;
[0040] S102: Preprocess the acquired SAR image data, including denoising, correction and enhancement;
[0041] S103: Generate corresponding target identification labels based on the preprocessed SAR image data.
[0042] Furthermore, the neural network model for target recognition includes a backbone network for feature extraction and a classifier composed of fully connected layers for classification;
[0043] The backbone network extracts high-level features from the input SAR image step by step through a series of convolutional layers;
[0044] The classifier maps the high-level features extracted by the backbone network to the final output category.
[0045] Furthermore, step S6 specifically includes the following steps:
[0046] S601: Select and prepare the test dataset;
[0047] S602: Load the trained global model parameters into the test environment;
[0048] S603: Perform inference on the global model in the test dataset to obtain the prediction result. The calculation expression for the prediction result is as follows:
[0049]
[0050] In the formula, For the predicted results, For global model parameters, x test For test data, f(·) is the model function;
[0051] S604: Compare the prediction results with the true labels y in the test dataset. test To compare the results, the validation accuracy Acc of the model is calculated. The expression for Acc is as follows:
[0052]
[0053] In the formula, D test For the test dataset, 1() is an indicator function that takes a value of 1 when the predicted result is the same as the true label, and 0 otherwise. For the i-th prediction result, y test,i For the i-th real label y test ;
[0054] S605: Evaluate model performance based on the model's validation accuracy (Acc);
[0055] S606: Optimize and improve the global model based on the model performance evaluation results.
[0056] Furthermore, step S605 also includes:
[0057] Record the training loss L for different numbers of clients train It is used to evaluate the learning process of the global model and the convergence speed under different client configurations;
[0058] Record the validation accuracy (Acc) of the global model in each communication round. round It is used to evaluate the changes and stability of model accuracy during training;
[0059] Record the global model validation accuracy (Acc) after completing the maximum number of communication rounds. final This is used to measure the overall effectiveness and performance of the training process with different numbers of clients.
[0060] This invention proposes a multi-client collaborative synthetic aperture radar (SAR) image target recognition method based on federated learning, aiming to address the dual challenges of insufficient SAR data sample quantity and data privacy protection. In traditional intelligent interpretation of SAR images, due to the high cost and difficulty of data acquisition, single satellite institutions often face the problem of insufficient data sample quantity, which severely limits the improvement of model performance and the expansion of application scope. Meanwhile, the sensitive geographical location and target information contained in SAR data makes data privacy protection crucial during data sharing.
[0061] Compared with the prior art, the present invention has the following advantages:
[0062] (1) This invention utilizes federated learning technology to allow multiple clients (such as different satellite agencies) to collaboratively train a model without sharing the original data. Specifically, each client trains the model locally using private data and uploads the updated model parameters to a central server for aggregation, forming new global model parameters. In this way, data is fully utilized to improve the model's generalization ability and recognition performance, while ensuring data privacy and security.
[0063] The innovation of this invention lies in its first application of federated learning to SAR image target recognition tasks, effectively solving the problems of insufficient SAR data samples and data privacy protection. By transferring model parameters among participants instead of raw data, multiple satellite agencies can jointly train the model without sharing data.
[0064] (2) Experimental results show that the multi-client collaborative SAR image target recognition method based on federated learning proposed in this invention effectively guarantees model performance under the premise of data privacy protection, and can gradually approach the performance of centralized training through iterative optimization. Specifically, the results are as follows:
[0065] 1. Ensure data privacy and security: Without sharing raw data, each client only transmits model parameters, effectively protecting data privacy and security.
[0066] 2. Ensuring Recognition Accuracy and Performance: Through multi-client collaborative training, the model can fully utilize data from each client. While ensuring data privacy, iterative optimization can gradually approach the performance of centralized training. In a task with 10 clients undergoing collaborative training, after 500 rounds of communication, the verification accuracy reached as high as 99.25%.
[0067] Therefore, this invention is of great significance in addressing the issues of insufficient data sample quantity and data privacy protection in SAR image target recognition. It effectively ensures model performance while protecting data privacy. This method has broad application prospects in military, civilian, and other fields. Attached Figure Description
[0068] Figure 1 This is a flowchart of a multi-client cooperative synthetic aperture radar (SAR) image target recognition method based on federated learning provided in an embodiment of the present invention;
[0069] Figure 2This is a schematic diagram of vehicle optical and SAR data in the MSTAR dataset provided in this embodiment of the invention, wherein (a) ZSU234; (b) ZIL131; (c) T72; (d) T62; (e) D7; (f) BTR70; (g) BTR60; (h) BRDM2; (i) BMP2; (j) 2S1;
[0070] Figure 3 This is a schematic diagram of a multi-client collaborative synthetic aperture radar (SAR) image target recognition task framework provided in an embodiment of the present invention;
[0071] Figure 4 This is a schematic diagram of a client-server network structure provided in an embodiment of the present invention;
[0072] Figure 5 This is a diagram showing the training loss results obtained using a federated learning algorithm under different numbers of client collaborative tasks, as provided in this embodiment of the invention.
[0073] Figure 6 This is a graph showing the verification accuracy results obtained using a federated learning algorithm under different numbers of client collaborative tasks, as provided in this embodiment of the invention. Detailed Implementation
[0074] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0075] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0076] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.
[0077] Example 1
[0078] This embodiment provides a multi-client collaborative SAR image target recognition method based on federated learning, which enables multiple satellite agencies to jointly train a model without sharing SAR data, thereby achieving data privacy protection and improving the performance of the target recognition model. The method includes the following steps:
[0079] S1: Collect SAR image data and preprocess it to generate corresponding target recognition labels, resulting in a labeled dataset. In this embodiment, the labeled dataset is further divided into a training set and a test set. The training set is used for the training process of the federated learning framework, and the test set is used for the testing process of the federated learning framework.
[0080] S2: Build a federated learning framework, which includes a server and multiple clients participating in collaborative training. Each client shares model parameters with the server; construct a neural network model for object recognition, initialize it, and distribute global model parameters to each client.
[0081] S3: Allocate local private data to each client based on the labeled dataset, and each client uses the local private data to train the local model and uploads the updated model parameters to the server;
[0082] S4: The server aggregates the model parameters uploaded by all clients, updates the global model parameters, and distributes the updated global model parameters to each client;
[0083] S5: Repeat steps S3 and S4 until the global model parameters reach convergence and the trained global model is obtained.
[0084] S6: Test the trained global model, obtain the validation accuracy, evaluate the model performance, and select the global model that passes the test for SAR image target recognition.
[0085] The following combination Figure 1 The flowchart of the multi-client cooperative synthetic aperture radar (SAR) image target recognition method based on federated learning in this embodiment is described.
[0086] Step S1, SAR image data acquisition and preprocessing: Acquire SAR image data, preprocess the data, generate labels, and divide the dataset into training and test sets.
[0087] Step S1 includes the following sub-steps:
[0088] Step S101: Data Acquisition. The satellite agency acquires SAR image data.
[0089] Step S102: Data preprocessing. The acquired SAR image data undergoes preprocessing operations such as denoising, correction, and enhancement to improve data quality.
[0090] Step S103: Generate labels. Generate corresponding target identification labels y based on the SAR image data.
[0091] Step S104: Split the dataset. In the example of collaborative multi-client SAR target recognition, we used the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset. The MSTAR dataset is a widely used SAR target recognition dataset containing SAR images of 10 types of military vehicles. Figure 2 This section presents schematic diagrams of optical and SAR images corresponding to various vehicle types in the MSTAR dataset. Table 1 shows the number of samples corresponding to the 10 target categories in the MSTAR dataset.
[0092] Each satellite agency will process its own labeled dataset D i Divided into training set D tr and test set D te In this example, the training and test sets consist of target data with pitch angles of 17° and 15°, respectively.
[0093] Specifically, in the collaborative multi-client SAR target recognition example, we assume that samples of each class are randomly and uniformly distributed among all clients, ensuring that there is no overlap or intersection of data between clients. The example simulates Independent and Identically Distributed (IID) scenarios, uniformly and randomly dividing the dataset among the clients participating in collaborative training. To verify the effectiveness and generalization of the proposed method, the example simulates configuration scenarios with 5 and 10 clients, and compares the distributed training results with centralized training (where all data is held by a single client).
[0094] Table 1 shows the number of samples for each of the 10 target classes in the MSTAR dataset.
[0095]
[0096] Step S2, Build a federated learning framework and initialize global model parameters: Build a federated learning framework and initialize the parameters of the global model for collaborative training.
[0097] Step S2 includes the following sub-steps:
[0098] Step S201: Establish a federated learning framework and determine the participating clients and servers. Specifically, this includes determining the clients U participating in collaborative training. ll The number of clients N and the central server S fl The multi-client collaborative SAR target recognition examples simulated configuration scenarios with 5 and 10 clients (i.e., N=5, 10), and compared the distributed training results with centralized training (all data is held by a single client, N=1).
[0099] Figure 3This presents a framework for multi-client collaborative synthetic aperture radar (SAR) image target recognition. The collaborative federated learning framework for multi-client SAR target recognition involves multiple SAR data clients that collaboratively train a global model without sharing raw data, thus protecting data privacy. Each client trains its model locally using its own data and only shares model parameters with a central server, which then aggregates these parameters to update the global model. The entire process is as follows: Figure 3 As shown, the process consists of three main steps: training task initialization, local model training and updating, and global model aggregation and updating. In this framework, blue arrows represent uploading local model parameters from the client to the server, and orange arrows represent distributing aggregated global model parameters from the server to the client. Specifically:
[0100] 1. Training task initialization: The server initializes the global model parameters and distributes them to each client.
[0101] 2. Local model training and updates: Each client trains the model locally using its private data and updates the local model parameters.
[0102] 3. Global model aggregation and update: The server receives local model parameters uploaded by all clients, aggregates them and updates the global model parameters, and then distributes the updated global model parameters to each client.
[0103] Step S202: Initialize global model parameters, setting initial weights and learning rate, etc. Specifically, this includes: defining the model architecture; initializing global model parameters w according to the preset neural network structure. g The initialized global model parameters are Set the learning rate η and other relevant hyperparameters, such as batch size B and number of training epochs E.
[0104] To effectively perform SAR target recognition in a federated learning environment, in this example, we designed a neural network model called ClientNet, such as... Figure 4 As shown, the model mainly consists of two parts: a backbone network for feature extraction and a classifier composed of fully connected layers for classification.
[0105] Figure 4The overall architecture of ClientNet is shown. The backbone network extracts high-level features from the input SAR image step by step through a series of convolutional layers, each followed by a batch normalization, ReLU activation, and pooling layer. This hierarchical approach ensures that the model can capture the spatial and hierarchical features of the input data. Specifically, the backbone network consists of four convolutional modules, as detailed in Table 2. After feature extraction, fully connected layers act as classifiers. The fully connected layers map the extracted features to the final output categories. Specifically, the fully connected layers consist of two layers. The first fully connected layer has 256 units, followed by a ReLU activation function and a Dropout layer to prevent overfitting. The second fully connected layer maps the 256 units to the number of output categories, and finally uses a Softmax activation function to convert the logits into multi-class classification probabilities.
[0106] Furthermore, in this example, the specific hyperparameter settings are as follows: batch size is 100, maximum communication rounds are 500, learning rate is 0.001, and the optimizer is AdamW. All experiments were performed on a single RTX 3090 GPU. Initializing the global model parameters through the above steps ensures the model's basic performance, providing a foundation for subsequent local training and global aggregation.
[0107] Table 2. Configuration of convolutional modules in the ClientNet backbone network
[0108]
[0109] Step S203: The server distributes the initial global model parameters. Server S fl initialize global model parameters Distributed to various client U fl .
[0110] Step S3: Local model training and update: Each client uses private data to train the local model and uploads the updated model parameters to the server.
[0111] Step S3 includes the following sub-steps:
[0112] Step S301: Each client loads the global model parameters. Each client U fl Load global model parameters in the current iteration t
[0113]
[0114] Step S302: Train the model using local private data. (Each client U)fl Using its private data D i The system undergoes several rounds of iterative training with the device, calculating the loss function and updating gradients until the local model parameters converge. (Each client U...) fl By minimizing the local loss function To update the local model parameters w i :
[0115]
[0116] Where η is the learning rate, l is the cross-entropy loss function, (x i,j ,y i,j () represents local data and corresponding labels.
[0117] Step S303: Upload the updated local model parameters to the server. After training is complete, each client U fl Update the local model parameters Send to server S fl .
[0118] Step S4: Global Model Aggregation and Update: The server aggregates all model parameters uploaded by clients and updates the global model parameters.
[0119] Step S4 includes the following sub-steps:
[0120] Step S401: Receive local model parameters uploaded by the client. Server S fl From various client U fl Receive transmitted local model parameters
[0121] Step S402: Aggregate local model parameters. Server U fl Based on a predetermined aggregation algorithm, such as the FedAvg algorithm, the model parameters of all clients are weighted averaged or otherwise aggregated to generate new global model parameters.
[0122]
[0123] Where N is the number of clients, |D i | represents the data volume of the i-th client. This represents the total amount of data.
[0124] Step S403: Update global model parameters. Server S fl Aggregated global model parameters Replace the old global model parameters:
[0125]
[0126] Step S404: Prepare for global model parameter download. The server prepares to download the new global model parameters. Distributed to various client U fl To begin a new round of training.
[0127] Step S5: Iterate the training until the global model converges.
[0128] Step S5 includes the following sub-steps:
[0129] Step S501: Distribute the updated global model parameters to each client. Server S fl New global model parameters Distributed to various client U fl .
[0130] Step S502: Each client repeats steps S301 and S302 to perform local model training and updates. Client U fl Use updated global model parameters Perform local training and upload the updated local model parameters to the server. Specifically, this includes: Client U fl Received new global model parameters Then, replace the local model parameters:
[0131]
[0132] Using local data D i Training is performed by minimizing the local loss function. Update local model parameters w i :
[0133]
[0134] After training, update the local model parameters. Send to server S fl .
[0135] Step S503: The server repeats steps S401 to S403 to perform global model aggregation and update. Server S fl Receive local model parameters uploaded by the client Then, a weighted average is performed using the FedAvg algorithm to generate new global model parameters.
[0136]
[0137] Step S504: Iteratively execute steps S5-1 to S5-3 until the global model parameter w is reached. gConvergence means that the model performance reaches the predetermined convergence criterion or the number of training epochs reaches the preset maximum value.
[0138] Step S6: Model testing and performance evaluation: Test the trained global model on the test dataset, obtain the validation accuracy, and evaluate the model performance.
[0139] Step S6 includes the following sub-steps:
[0140] Step S601: Prepare the test dataset. Select and prepare an independent test dataset D, separate from the training dataset. te This is to ensure the objectivity of the test results.
[0141] Step S602: Load global model parameters. Load the finally converged global model parameters. Load it into the test environment.
[0142] Step S603: Perform model inference on the test dataset. Use the test dataset D. te Perform inference on the global model to obtain the model's prediction results.
[0143]
[0144] Where, x test For the test data, f(·) is the model function.
[0145] Step S604: Calculate the validation accuracy. Use the model's prediction results... Compared with the true label y of the test dataset test By comparison, the validation accuracy (Acc) of the model is calculated. val To evaluate the model's recognition performance:
[0146]
[0147] Here, 1() is an indicator function, which takes the value 1 when the prediction result is the same as the true label, and 0 otherwise.
[0148] Step S605: Evaluate model performance. In addition to verifying accuracy, other performance metrics can be calculated to comprehensively evaluate the model and determine its effectiveness and reliability in practical applications. Specifically:
[0149] 1. Record the training loss L for different numbers of clients. train This metric provides information about the model's learning process and convergence speed under different client configurations:
[0150]
[0151] Figure 5The graph illustrates the changes in training loss over 500 rounds of communication with different numbers of clients. As the number of communication rounds increases, the training loss gradually decreases and stabilizes, indicating that the model is gradually converging. Specifically, the training loss decreases most rapidly when the number of clients is 1 (centralized training). When the number of clients is 5 (federated learning), the curve is similar to that of centralized training. When the number of clients is 10 (federated learning), the rate of decrease in training loss is slightly slower, but it eventually converges as well. This shows that although the training loss of federated learning fluctuates slightly in the initial stages, it gradually decreases with increasing communication rounds, eventually approaching the effect of centralized training.
[0152] 2. Record the validation accuracy of the global model in each communication round. This helps in understanding how model accuracy changes and remains stable during training.
[0153] Figure 6 The graph shows the changes in validation accuracy over 500 rounds of communication with different numbers of clients. As can be seen from the graph, the model's validation accuracy gradually improves and stabilizes with increasing communication rounds, indicating that the model is gradually converging. The validation accuracy improves most rapidly when the number of clients is 1 (centralized training), while the validation accuracy fluctuates significantly in the early stages when the number of clients is 5 and 10 (federated learning), but eventually stabilizes and approaches the effect of centralized training.
[0154] 3. Record the global model validation accuracy after completing the maximum number of communication rounds (e.g., 500 rounds). Measure the overall effectiveness and performance of the training process with different numbers of clients (e.g., 1, 5, and 10).
[0155] Table 3 shows the validation accuracy of the model after 500 rounds of communication under different client number configurations. Experimental results show that centralized training (N=1) achieves a validation accuracy of 99.26%, while federated learning achieves 99.17% and 99.25% with 5 and 10 clients, respectively. These results demonstrate that the proposed scheme achieves collaborative training across multiple clients while maintaining high model performance.
[0156] Table 3. Validation accuracy of the model under different client number configurations (500 rounds of communication)
[0157]
[0158] Step S606: Optimize the model. Based on the performance evaluation results, optimize and improve the model to enhance its performance and generalization ability. Specific optimization measures that can be taken are as follows:
[0159] 1. Adjust hyperparameters:
[0160] • Learning rate: Adjust the learning rate appropriately based on the model's convergence during training. If the model converges too slowly, increase the learning rate; if the model oscillates or overfits during training, decrease the learning rate.
[0161] • Batch size: Adjust the batch size based on the model's training efficiency and memory usage. A larger batch size can speed up training but may lead to insufficient memory; a smaller batch size can improve the model's generalization ability but slows down training.
[0162] • Communication Rounds: Adjust the maximum number of communication rounds based on the model's performance in different communication rounds. Experimental results show that the model gradually converges after 500 communication rounds, and the number of communication rounds can be appropriately increased or decreased according to specific needs.
[0163] 2. Improve the model architecture:
[0164] • Increase network depth or width: Improve the model's feature extraction capabilities by increasing the number of convolutional layers or filters.
[0165] • Introduce other neural network modules: Use more advanced neural network modules (such as residual networks, attention mechanisms, etc.) to improve model performance.
[0166] 3. Optimize training strategies:
[0167] • Use more advanced optimization algorithms, such as AdamW and RMSprop, to accelerate the model's convergence process.
[0168] • Early stopping strategy: During training, stop training early when the validation accuracy no longer improves significantly to prevent overfitting.
[0169] • Regularization methods: Introduce L2 regularization or Dropout techniques to prevent model overfitting and improve its generalization ability.
[0170] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.
Claims
1. A multi-client cooperative SAR image target recognition method based on federated learning, characterized in that, Includes the following steps: S1: Acquire SAR image data and preprocess it to generate corresponding target recognition labels, thus obtaining a labeled dataset; S2: Build a federated learning framework, which includes a server and multiple clients participating in collaborative training. Each client shares model parameters with the server; construct a neural network model for object recognition, initialize it, and distribute global model parameters to each client. S3: Allocate local private data to each client based on the labeled dataset, and each client uses the local private data to train the local model and uploads the updated model parameters to the server; S4: The server aggregates the model parameters uploaded by all clients, updates the global model parameters, and distributes the updated global model parameters to each client; S5: Repeat steps S3 and S4 until the global model parameters reach convergence and the trained global model is obtained. S6: Test the trained global model and select the global model that passes the test for SAR image target recognition; Step S2 specifically includes the following steps: S201: Establish a semi-decentralized federated learning framework, determine the participating clients and servers, as well as the number of clients, and each client shares model parameters with the server, which aggregates the model parameters of each client to update the global model; S202: Pre-build a neural network model for target recognition and initialize global model parameters; S203: The server distributes the initialized global model parameters to each client; The neural network model for target recognition includes a backbone network for feature extraction and a classifier consisting of fully connected layers for classification. The backbone network extracts high-level features from the input SAR image step by step through a series of convolutional layers; The classifier maps the high-level features extracted by the backbone network to the final output category. Step S3 specifically includes the following steps: S301: Each client loads the global model parameters transmitted by the server; S302: Each client holds a local dataset allocated based on the labeled dataset. Each client performs local training based on the allocated global model parameters, calculates the loss function, and updates the gradient. The corresponding update expression for the local model parameters is: In the formula, These are the local model parameters at the (t+1)th iteration. These are the local model parameters at the t-th iteration. For learning rate, For model parameters The result of the loss function calculation is as follows: Let cross-entropy be the loss function. For the data of the i-th client, For local data and corresponding target identification labels; S303: Each client sends the updated local model parameters to the server.
2. The multi-client cooperative SAR image target recognition method based on federated learning according to claim 1, characterized in that, Step S4 specifically includes the following steps: S401: The server receives local model parameters transmitted from each client; S402: The server performs aggregation operations on the local model parameters of each client according to a predetermined aggregation algorithm to generate new global model parameters; S403: The server updates the global model parameters based on the new global model parameters; S404: The server distributes the updated global model parameters to each client.
3. The multi-client cooperative SAR image target recognition method based on federated learning according to claim 2, characterized in that, The aggregation algorithm is the FedAvg algorithm, which is used to perform a weighted average of the model parameters of all clients. The calculation expression of the FedAvg algorithm is as follows: In the formula, Here are the global model parameters for the (t+1)th iteration, and N is the number of clients. Let i be the amount of data from the i-th client. The total amount of data, These are the local model parameters for the i-th client during the t-th iteration.
4. The multi-client cooperative SAR image target recognition method based on federated learning according to claim 1, characterized in that, Step S5 specifically includes the following steps: S501: The server distributes the updated global model parameters to each client; S502: Each client executes step S3 to train the local model; S503: The server executes step S4 to aggregate and update the global model; S504: Iteratively execute steps S501-S503 until the global model parameters converge.
5. The multi-client cooperative SAR image target recognition method based on federated learning according to claim 1, characterized in that, Step S1 specifically includes the following steps: S101: Acquire SAR image data via satellite; S102: Preprocess the acquired SAR image data, including denoising, correction and enhancement; S103: Generate corresponding target identification labels based on the preprocessed SAR image data.
6. The multi-client cooperative SAR image target recognition method based on federated learning according to claim 1, characterized in that, Step S6 specifically includes the following steps: S601: Select and prepare the test dataset; S602: Load the trained global model parameters into the test environment; S603: Perform inference on the global model in the test dataset to obtain the prediction result. The calculation expression for the prediction result is as follows: In the formula, For the predicted results, These are global model parameters. For test data, For model functions; S604: Compare the prediction results with the true labels of the test dataset. Compare the results and calculate the validation accuracy of the models. The accuracy The calculation expression is: In the formula, For the test dataset, This is an indicator function; it takes a value of 1 when the predicted result matches the true label, and 0 otherwise. For the i-th prediction result, For the i-th real label ; S605: Based on the model's validation accuracy Evaluate model performance; S606: Optimize and improve the global model based on the model performance evaluation results.
7. The multi-client cooperative SAR image target recognition method based on federated learning according to claim 6, characterized in that, Step S605 also includes: Record the training loss for different numbers of clients It is used to evaluate the learning process of the global model and the convergence speed under different client configurations; Record the validation accuracy of the global model in each communication round. It is used to evaluate the changes and stability of model accuracy during training; Record the global model validation accuracy after completing the maximum number of communication rounds. This is used to measure the overall effectiveness and performance of the training process with different numbers of clients.