In recent years, ship target detection in synthetic aperture radar (SAR) images has significantly progressed due to the rapid development of deep learning (DL). However, since only the spatial feature information of ship targets is utilized, the current DL-based SAR ship detection approaches cannot achieve a satisfactory performance, especially in the case of multiscale, rotations, or complex backgrounds. To address these issues, in this paper, a novel deep-learning network for SAR ship rotation detection, called a morphology and topology-based feature alignment network, is proposed which can better exploit the morphological features and inherent topological structure information. This network consists of the following three main steps: First, deformable convolution is introduced to improve the representational ability for irregularly shaped ship targets, and subsequently, a morphology and topology feature pyramid network is developed to extract inherent topological structure information. Second, based on the aforementioned features, a rotation alignment feature head is devised for fine-grained processing as well as aligning and distinguishing the features; to enable regression prediction of rotated bounding boxes; and to adopt a parameter-sharing mechanism to improve detection efficiency. Therefore, utilizing morphological and inherent topological structural information enables a superior detection performance to be achieved. Finally, we evaluate the effectiveness of the proposed method using the rotated ship detection dataset in SAR images (RSDD-SAR). Our method outperforms other DL-based algorithms with fewer parameters. The overall average precision is 90.84% and recall is 92.21%. In inshore and offshore scenarios, our method performs well for the detection of multi-scale and rotation-varying ship targets, with its average precision reaching 66.87% and 95.72%, respectively.

This paper addresses the problem of weak maneuvering target detection in the space-based bistatic radar system through long-time coherent integration (LTCI). The space-based bistatic radar is vulnerable to the high-order range migration (RM) and Doppler frequency migration (DFM), since the target, the receiver and the transmitter all can play fast movement independently. To correct high- order RM and DFM, this usually involves joint high-dimensional parameter searching, incurring a large computational burden. In our previous work, a dual-scale (DS) decomposition of motion parameters was proposed, in which the optimal GRFT is conditionally decoupled into two cascade procedures called the modified generalized inverse Fourier transform (GIFT) and generalized Fourier transform (GFT), resulting in the DS-GRFT detector. However, even if the DS-GRFT detector preserves the superior performance and dramatically decreases the complexity, high-dimensional searching is still required. In this paper, by analyzing the structure of the DS-GRFT detector, we further designed a conditioned cubic phase function (CCPF) tailored to the range–slow-time signal after GIFT, breaking the joint high-dimensional searching into independent one-dimensional searching. Then, by connecting the proposed CCPF with the GIFT, we achieved a new LTCI detector called the DS-GIFT-CCPF detector, which obtained a significant computational cost reduction with acceptable performance loss, as demonstrated in numerical experiments.