The traditional synthetic aperture radar (SAR) imaging algorithm typically demands substantial computational resources due to its involvement in large-scale matrix operations, which presents challenges for imaging in expansive scenes. By leveraging the reversibility of the imaging process, a method that simulates the inverse process of generating echoes can be employed to expedite imaging. This approach shares the computational complexity of the matched filter algorithm, thereby enabling imaging of large scenes. However, the iteration-based algorithm necessitates manual parameter adjustments, which can be somewhat arbitrary and challenging to optimize effectively. Consequently, employing neural networks to train these parameters not only enhances imaging speed but also enhances imaging quality. Moreover, the Self-supervised Azimuth Ambiguity Suppression Network (SAAS-Net) introduced in this study effectively achieves azimuth ambiguity suppression without necessitating alterations to the hardware architecture. The experiments indicate that the algorithm proposed in this paper can achieve rapid imaging while maintaining computational accuracy consistent with matched filter algorithms such as chirp-scaling algorithm.

Synthetic Aperture Radar (SAR) has emerged as a critical technology for detecting and classifying objects such as ships in challenging environments. However, few-shot learning remains a challenge due to the limited availability of labeled SAR data, complex radar backscatter, and variations in imaging parameters. In this paper, we propose a novel network, Scattering Point Topology for Few-Shot Ship Classification (SPT-FSC), which addresses these challenges by incorporating scattering characteristics into the network learning process through a scattering point topology (SPT) based on scattering key points. We design a topology encoding branch (TEB) through a series of operations to encode the topological information of scattering points, resulting in a scattering point topology embedding (SPTE) that improves the network’s adaptability to the imaging mechanism and reduces imaging variability in SAR images. To effectively fuse the SPTE and image features extracted from a convolutional neural network (CNN), we introduce a novel mechanism called reciprocal feature fusion attention (RFFA). Furthermore, to address the limited diversity in the training data, we apply transfer learning methodologies and construct a fine-grained ship classification dataset by combining the OpenSARShip and FUSAR-Ship datasets. Our comprehensive experiments on these datasets demonstrate the effectiveness of our proposed SPT-FSC method, achieving high accuracy and robustness in few-shot ship classification tasks for SAR images, outperforming existing methods in this domain.