Ship recognition in synthetic aperture radar (SAR) images is a significant and fundamental step in the maritime surveillance. However, recognition of ships inevitably faces background interference in the maritime environment. The interference guides the network focusing on useless even harmful regions. To deal with issue, a mixed attention mechanism consists of coordinate and Squeeze-and-Excitation(SE) attentions is introduced. The mixed attention can guide the network to focus more on the target region, decreasing the influence of useless interference regions. Experimental and visualize results on benchmark dataset OpenSARShip validate the effectiveness of our idea.

With the launch of high-resolution (HR) synthetic aperture radar (SAR) imaging satellites and the rapid development of convolutional neural networks (CNNs), ship recognition in HR SAR images has shown further improvements. Unlike ship targets in low and medium-resolution SAR images, which only possess a few pixels and present a spot-like appearance, ship targets in HR SAR images pose a larger area of pixels. However, CNN lacks the power to model dependencies between long-range features occupying large areas of pixels. A convolutional transformer (CvT) is introduced to deal with this issue. CvT integrates the local features of CNNs and the capability of capturing long-range dependencies of transformers to model both local and global dependencies for ship recognition in an efficient way. The cosine-margin loss is also applied to constraint strictly the distribution of the features to further improve the performance. Experimental results on the benchmark FUSAR-Ship dataset demonstrate the effectiveness of the proposed method for ship ATR in HR SAR images.

Synthetic aperture radar (SAR) has been widely applied in maritime target detection. However, most existing SAR ship detection algorithms based on convolutional neural network (CNN) only use single polarization SAR images for detection, neglecting to further improve the detection performance by utilizing the rich polarization information of the SAR images. To deal with this issue, this paper proposes a Cascaded Feature Fusion Pyramid Network (CFFPN) for ship detection in dual-polarization SAR images. The CFFPN builds a cascaded feature fusion module (CFFM) to fuse the enriched polarization information in SAR images. Extensive evaluations conducted on the the dual-polarization SAR ship detection dataset showcase the remarkable effectiveness of CFFPN, achieving an average precision (AP) of 93.4%. This outperforms the other five competitive methods. Notably, CFFPN exhibits a notable improvement of 1.3% in AP compared to the second-best method.

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.

In the field of maritime surveillance, ship recognition in synthetic aperture radar (SAR) images is a significant and fundamental step. Compared with traditional methods, convolutional neural networks (CNNs) tend to be the mainstream in SAR ship recognition. However, these methods ignore one core issue. Multi-scale features can enhance the expression ability of features, which are currently not well-exploited. In response to this problem, a novel feature weaving pyramid network (FWPN-Net) is proposed. FWPN-Net contains a multi-scale feature weaving module (MFWM), which can integrate high level semantic information and low level detailed information to obtain better representations of multi-scale SAR ship features. Experimental results on benchmark dataset OpenSARShip show that the proposed FWPN-Net performs better than classic CNN methods and modern SAR ship recognition CNN method.

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.

In the field of remote sensing, harbor detection in SAR images has an important application prospect. However, the complex coastline of SAR images increases the difficulty of harbor detection. In response to this problem, a cascaded harbor detection (CHD) method for SAR image based on corner and coastline features is proposed in this paper. First, coast-line is extracted from SAR image by sea-land segmentation. Then, in the first step rough detection, corner detection is performed on the coastline and the detected corners are automatically clustered to locate the harbor candidate areas. Finally, the second step precise detection is carried out on the coast-line of harbor candidate areas, where coastline feature detection is completed by using corners again to remove the fake harbor targets in harbor candidate areas. Experimental results based on satellite-borne SAR data prove the proposed CHD method enjoys a preferable detection performance compared with existing harbor detection methods.