Ship recognition in synthetic aperture radar (SAR) images is essential for many applications in maritime surveillance tasks. Recently, convolutional neural network (CNN)-based methods tend to be the mainstream in SAR recognition. Though considerable developments have been achieved, there are still several challenging issues toward superior ship recognition performance: 1) Ships have a large variance in size, making it difficult to recognize ships by using a single scale features of CNN. 2) The SAR ship’s large aspect ratio presents an obvious geometric characteristic. However, standard convolution is limited by the fixed convolution kernel, which is less effective in processing elongated SAR ships. 3) Existing CNN classifiers with softmax loss are less powerful to deal with intraclass diversity and interclass similarity in SAR ships. In this paper, we propose a task-specific hierarchically designed network with a spherical space classifier (HDSS-Net) to alleviate the above issues. Firstly, to realize SAR ship recognition with large size variation, a feature aggregation module (FAM) is designed for obtaining a feature pyramid that has strong representational power at all scales. Secondly, a FeatureBoost module (FBM) is devised to provide rectangular receptive fields to refine the features generated by FAM. Finally, a novel spherical space classifier (SSC) is proposed to expand the interclass margin and compress the intraclass feature distribution by fully taking advantage of the property of spherical space. The experimental results on two benchmark datasets (OpenSARShip and FUSAR-Ship) jointly show that the proposed HDSS-Net performs better than classic CNN methods and novel SAR ship recognition CNN methods.

Maritime surveillance is extensively concerned by worldwide authorities, in which ship recognition in synthetic aperture radar (SAR) images is a significant and fundamental component. Though some development has been achieved in the SAR ship recognition task, two areas remain inadequately explored, which are the comprehensive utilization of multi-scale features and the deployment of the prior knowledge of the ship shape. In this paper, a novel ship augmented attention network (SA ² Net) for ship recognition is proposed, which comprehensively utilizes the multi-scale features and integrates the ship shape prior to the end-to-end network. On one hand, due to the unequal effects of different scales, a scale attention module (SAM) is proposed to adaptively select and assign weights to desired feature scales while disregarding irrelevant scales. Moreover, a feature weaving module (FWM) is constructed to merge semantic and detailed features produced by the high-to-low backbone, enriching representations across all scales of ship targets. On the other hand, in order to incorporate the priory knowledge of the ship shape into the network, we develop a feature augmentation module (FAM) to further boost the ship recognition accuracy. This module can provide rectangular receptive fields that align with the shape of ships, wherein a limitation encountered with traditional square convolutions. Comprehensive experiments on representative three- and six-category OpenSARShip tasks and seven-category FUSAR-Ship tasks show that our SA ² Net demonstrates superior performance when compared to the current state-of-the-art methods.

Automatic target recognition (ATR) holds a crucial position in synthetic aperture radar (SAR) image interpretation. Despite deep learning advancements have significantly propelled SAR ATR, addressing the challenge of target recognition with a few training data remains a vital concern in SAR applications. Two main issues still exist: 1) In few-shot SAR ATR, the depth and width of CNN-based models are limited, which restricts its modeling capacity, and thus extracting discriminative generalized features remains challenging. 2) With only a few labeled SAR images, the resultant class distribution is biased due to the intra-class diversity and inter-class similarity of SAR samples, which degrades the recognition performance. To address these challenges, in this letter, we propose a novel dynamically weighted prototypical learning (DWPL) method. Firstly, to extract discriminative generalized features from SAR images, we propose a new convolutional transformer network with great capacity to capture long-range dependencies of local features, together with an effective random task augmentation strategy. Secondly, in consideration of intra-class diversity and inter-class similarity, a dynamically weighted prototypical module (DWPM) is designed to adaptively assign weights to the few labeled samples that have varying discriminative information. This enables the model to effectively explore the hidden features in few samples. Through experiments conducted on the moving and stationary target acquisition and recognition (MSTAR) dataset, our method achieves recognition accuracies of 97.22% and 92.01% for 3-way 5-shot and 3-way 1-shot SAR ATR tasks in SOC, revealing significant and robust recognition performance.

During the no-load closing process, transformer will produce a magnetizing inrush current which will make the transformer relay malfunction and make adversely affect to the transformer equipment and power grid. Relative to a variety of external measures limiting the inrush current, the phase selection switch technology is more effective to reduce inrush current. The author analyzed inrush current by theory and dynamic simulation experiment. By controlling the closing angle to be less than 0° to 72° of the opening angle, inrush current is well suppressed.

Although deep learning-based approaches have significantly advanced intelligent radar image interpretation, existing methods predominantly rely on single-view inputs, which inherently limits target recognition performance. In practical scenarios, multi-view radar imagery provides richer structural and feature representations of targets compared to single-view observations. To effectively exploit this comprehensive information, we propose a new Multi-view Laplacian Propagation Fusion Network (MLPFN) for radar image target recognition. First, a Reparameterization Perspective-Aware (RPA) backbone is designed, incorporating a Dual-level Multi-path Rep-Kernel (DMPK) and a Multi-level Dual-path Attention Module (MDAM). This architecture mitigates the sensitivity of recognition performance to aspect angle variations and enhances the extraction of discriminative target features. Second, a Laplacian Multi-view Feature Propagation Algorithm (LMFPA) is introduced to encode structural correlations among different views, thereby augmenting the feature discriminability for multi-view targets. Third, a Multi-view Feature Adaptive Fusion Module (MFAFM) is developed to integrate global contextual information across views into a unified representation, reducing redundancy while maximizing feature distinctiveness. Furthermore, a Stage-wise Guided Constraint Training Mechanism (SGCTM) is devised to supervise the feature extraction process across different network layers during training, refining the network’s feature learning and recognition capabilities. Extensive experiments conducted on multi-view radar image datasets demonstrate the effectiveness of the proposed MLPFN.