Synthetic aperture radar (SAR) image classification based on deep learning methods has attracted huge enthusiasm among researchers. Sufficient training data guarantee the generalization ability of neural networks. However, it is quite difficult to collect large amount of high-quality and labeled SAR data in real situations. In this letter, a method based on contrastive learning and pseudo-labels is proposed to solve the few-shot SAR image classification problem. First, a siamese structure is introduced to learn semantic representations of SAR target images. The feature vectors obtained in this stage can reveal the similarity between bi-temporal SAR images. Then, the backbone network of the siamese network is connected to a classification head for instance-level classification. The pseudo-label strategy is adopted to gradually improve the performance. A loss function that changes with iteration is designed to optimize the training process with true labels and pseudo-labels. Besides, considering the imprecision of single-network architecture, we adopt dual-networks and cross-training strategies. The final result is obtained by integrating the results of the dual-networks. Ablation studies show that the designed components of the proposed method are effective. Comparative experiments on the moving and stationary target acquisition and recognition (MSTAR) data set demonstrate the superior performance of the proposed method when dealing with limited training data in the SAR classification.

Since the start of the Sentinel-1 (S-1) mission, numerous cases of severe image degradation caused by radio frequency interference (RFI) have been reported, which puts forward an urgent need for RFI identification and mitigation. In this article, an automatic RFI identification method is proposed based on a Siamese-type deep convolutional neural network (Siam-CNN-RIM). The Siam-CNN-RIM can be served as a preprocessing step before RFI mitigation to identify whether an S-1 image is RFI-contaminated or not. Different from traditional RFI identification networks which only use a single image as input, an additional image in the repeat-pass time series is also fed into the input of Siam-CNN-RIM as a reference. Both the input images correspond to the same illuminated area and pass through the same convolutional layer followed by an energy function such that the different features caused by RFI can be extracted and the background terrain features can be ignored. This is beneficial for distinguishing the real RFI signatures and the similar terrain signatures that may cause false positives and thus improving the RFI identification performance. Experimental results show that the proposed method is robust in different scenarios and can achieve more than 97% RFI identification accuracy, even for the open-set task where the test scenarios are not included in the training set.

Change detection, as an essential part of synthetic aperture radar (SAR) automatic target recognition (ATR) systems, remains a challenging problem. The lack of labeled data seriously impedes the development of deep learning-based methods in SAR applications. In this letter, we propose a patch-based semisupervised method to detect changed pixels from limited training data in the field of SAR. The complete approach includes the unsupervised pretraining stage and the iterative discrimination stage. The states of pixels are determined by comparing the corresponding image patches. First, to reduce the impact of insufficient data, we train a trapezium U-Net-like structure to extract representative and generalized features by unsupervised training on all pixels. Feature maps contain detailed and semantic information due to repetitive feature fusions. A designed feature activation module is utilized to recalibrate fused features. Finally, the discrimination is finished by a simple classification convolutional neural network (CNN) following a two-step training strategy. Ablation studies indicate the function provided by the proposed modifications. Experiments on real SAR images demonstrate that the proposed method can improve the detection accuracy by more than 1.2% compared with other state-of-the-art deep learning-based methods.

For parallel bistatic forward-looking synthetic aperture radar (SAR) imaging, the instantaneous slant range is a double-square-root expression due to the separate transmitter-receiver system form. The hyperbolic approximation provides a feasible solution to convert the dual square-root expression into a single-square-root expression. However, some high-order terms of the range Taylor expansion have not been considered during the slant range approximation procedure in existing methods, and therefore, inaccurate phase compensation occurs. To obtain a more accurate compensation result, an improved hyperbolic approximation range form with high-order terms is proposed. Then, a modified omega-K algorithm based on the new slant range form is adopted for parallel bistatic forward-looking SAR imaging. Several simulation results validate the effectiveness of the proposed imaging algorithm.