High-resolution range profile (HRRP) open-set recognition (OSR) consists of in-distribution (ID) classification and out-of-distribution (OOD) rejection. While ID classification is straightforward, overconfidence in recognition leads to failed OOD rejection because of the absence of OOD samples and constrained recognition performance during training. Overconfidence can be mitigated by generating pseudo-OOD samples and integrating prior knowledge as additional information to support OSR. However, appropriate pseudo-OOD samples should differ from ID classes while remaining within the HRRP domain. Integrating prior knowledge may cause confusion between ID and near-OOD samples with similar priors. To address this, a physics-guided HRRP OSR framework (PGR) is proposed in this paper, consisting of three modules: physics-guided pseudo-OOD generation (PO), prior-decoupled feature extraction (PF), and OOD feature loss (OFL). PO generates appropriate pseudo-OOD HRRPs based on the relationship between radial and physical scales. PF integrates prior knowledge while decoupling similar priors from OOD rejection by assigning virtual priors. OFL robustly separates pseudo-OOD and ID features using a contrastive loss. Experiments on measured and simulated HRRP datasets validate the superiority of PGR over state-of-the-art methods in far-OOD and near-OOD scenarios.

In recent years, research on knowledge graphs has exploded due to their capability of effective organization and representation for massive heterogeneous data. However, existing knowledge graphs are often incomplete and contain incorrect triples, which easily incurs a negative impact on the performance of downstream tasks. Besides, few works have been done on the construction of ship knowledge graph. For the former, the mainstream solution is link prediction, also known as knowledge graph completion. To this end, we propose a novel method of knowledge graph embeddings for link prediction, called QuatGAT, combining graph attention mechanism with quaternion embeddings. Specifically, multi-head attention mechanism is employed firstly to obtain entity embeddings by capturing features of both entities and relations of the neighborhood. Then, to represent relations more sufficiently, we use quaternion embeddings to explicitly represent relations as well as entities. Experimental results on benchmark datasets FB15k and FB15k-237 demonstrate the superiority of QuatGAT over existing state-of-the-art methods. Moreover, in terms of ship knowledge graph construction, we also build a multimodal ship knowledge graph named MSKG. Likewise, experimental results on this dataset verify the effectiveness of QuatGAT.

High-resolution range profile (HRRP) target recognition has garnered significant attention in radar automatic target recognition (RATR) research for its rich structural information and low computational costs. With the rapid advancements in deep learning, methods for HRRP target recognition that leverage deep neural networks have emerged as the dominant approaches. Nevertheless, these traditional closed-set recognition methods require labeled data for every class in training, while in reality, seen classes and unseen classes coexist. Therefore, it is necessary to explore methods that can identify both seen and unseen classes simultaneously. To this end, a semantic- and physical-guided generative network (SPGGN) was innovatively proposed for HRRP generalized zero-shot recognition; it combines a constructed knowledge graph with attribute vectors to comprehensively represent semantics and reconstructs strong scattering points to introduce physical constraints. Specifically, to boost the robustness, we reconstructed the strong scattering points from deep features of HRRPs, where class-aware contrastive learning in the middle layer effectively mitigates the influence of target-aspect variations. In the classification stage, discriminative features are produced through attention-based feature fusion to capture multi-faceted information, while the design of balancing loss abates the bias towards seen classes. Experiments on two measured aircraft HRRP datasets validated the superior recognition performance of our method.

In the context of Synthetic Aperture Radar (SAR) image recognition, traditional methods often struggle with the intrinsic limitations of SAR data, such as weak texture, high noise, and ambiguous object boundaries. This work explores a novel perspective by reformulating SAR target recognition as a multimodal reasoning task. We leverage multimodal large language models (MLLMs), specifically GPT-4o, to perform target classification based on SAR imagery, guided by candidate categories and enhanced with Chain-of-Thought (CoT) reasoning. A new dataset is constructed based on the FAIR-CSAR benchmark, comprising raw SAR images, structured target annotations, candidate label sets, and GPT-generated CoT reasoning chains. Experimental results show that the MLLMs are capable of generating logically coherent and interpretable inferences in most scenarios. Our analysis highlights both the strengths and current limitations of MLLMs in interpreting SAR imagery, and we provide detailed insights into model behavior through failure case analysis. This work demonstrates the feasibility of incorporating MLLMs into SAR analysis pipelines and establishes a foundation for future research in SAR-oriented visual reasoning.