At present, deep learning has been widely used in SAR ship target detection, but the accurate and real-time detection of multi-scale targets still faces tough challenges. CNN-based SAR ship detectors are challenged to meet real-time requirements because of a large number of parameters. In this paper, we propose a lightweight, single-stage SAR ship target detection model called YOLO-based lightweight multi-scale ship detector (LMSD-YOLO), with better multi-scale adaptation capabilities. The proposed LMSD-YOLO consists of depthwise separable convolution, batch normalization and activate or not (ACON) activation function (DBA) module, Mobilenet with stem block (S-Mobilenet) backbone module, depthwise adaptively spatial feature fusion (DSASFF) neck module and SCYLLA-IoU (SIoU) loss function. Firstly, the DBA module is proposed as a general lightweight convolution unit to construct the whole lightweight model. Secondly, the improved S-Mobilenet module is designed as the backbone feature extraction network to enhance feature extraction ability without adding additional calculations. Then, the DSASFF module is proposed to achieve adaptive fusion of multi-scale features with fewer parameters. Finally, the SIoU is used as the loss function to accelerate model convergence and improve detection accuracy. The effectiveness of the LMSD-YOLO is validated on the SSDD, HRSID and GFSDD datasets, respectively, and the experimental results show that our proposed model has a smaller model volume and higher detection accuracy, and can accurately detect multi-scale targets in more complex scenes. The model volume of LMSD-YOLO is only 7.6MB (52.77% of model size of YOLOv5s), the detection speed on the NVIDIA AGX Xavier development board reached 68.3 FPS (32.7 FPS higher than YOLOv5s detector), indicating that the LMSD-YOLO can be easily deployed to the mobile platform for real-time application.

Current Synthetic Aperture Radar (SAR) image object detection methods require huge amounts of annotated data and can only detect the categories that appears in the training set. Due to the lack of training samples in the real applications, the performance decreases sharply on rare categories, which largely inhibits the detection model from reaching robustness. To tackle this problem, a novel few-shot SAR object detection framework is proposed, which is built upon the meta-learning architecture and aims at detecting objects of unseen classes given only a few annotated examples. Observing the quality of support features determines the performance of the few-shot object detection task, we propose an attention mechanism to highlight class-specific features while softening the irrelevant background information. Considering the variation between different support images, we also employ a support-guided module to enhance query features, thus generating high-qualified proposals more relevant to support images. To further exploit the relevance between support and query images, which is ignored in single class representation, a dynamic relationship learning paradigm is designed via constructing a graph convolutional network and imposing orthogonality constraint in hidden feature space, which both make features from the same category more closer and those from different classes more separable. Comprehensive experiments have been completed on the self-constructed SAR multi-class object detection dataset, which demonstrate the effectiveness of our few-shot object detection framework in learning more generalized features to both enhance the performance on novel classes and maintain the performance on base classes.

Nearshore ship detection has significant applications in both the military and civilian domains. Compared to synthetic aperture radar (SAR), polarimetric synthetic aperture radar (PolSAR) provides richer information for analyzing the scattering mechanisms of ships and enables better detection of ship targets. However, ships in nearshore areas tend to be highly concentrated, and ship detection is often affected by adjacent strong scattering, resulting in false alarms or missed detections. While the GP-PNF detector performs well in PolSAR ship detection, it cannot obtain satisfactory results in these scenarios, and it also struggles in the presence of azimuthal ambiguity or strong clutter interference. To address these challenges, we propose a nearshore ship detection method named ECD-PNF by integrating superpixel-level GP-PNF and refined polarimetric decomposition. Firstly, polarimetric superpixel segmentation and sea–land segmentation are performed to reduce the influence of land on ship detection. To estimate the sea clutter more accurately, an automatic censoring (AC) mechanism combined with superpixels is used to select the sea clutter superpixels. By utilizing refined eight-component polarimetric decomposition to improve the scattering vector, the physical interpretability of the detector is enhanced. Additionally, the expression of polarimetric coherence is improved to enhance the target clutter ratio (TCR). Finally, this paper combines the third eigenvalue of eigenvalue–eigenvector decomposition to reduce the impact of azimuthal ambiguity. Three spaceborne PolSAR datasets from Radarsat-2 and GF-3 are adopted in the experiments for comparison. The proposed ECD-PNF method achieves the highest figure of merit (FoM) value of 0.980, 1.000, and 1.000 for three datasets, validating the effectiveness of the proposed method.

The availability of high-quality and ample synthetic aperture radar (SAR) image datasets is crucial for understanding and recognizing target characteristics. However, in practical applications, the limited availability of SAR target images significantly impedes the advancement of SAR interpretation methodologies. In this study, we introduce a Generative Adversarial Network (GAN)-based approach designed to manipulate the target azimuth angle with few samples, thereby generating high-quality target images with adjustable angle ranges. The proposed method consists of three modules: a generative fusion local module conditioned on image features, a controllable angle generation module based on sparse representation, and an angle discrimination module based on scattering point extraction. Consequently, the generative modules fuse semantically aligned features from different images to produce diverse SAR samples, whereas the angle synthesis module constructs target images within a specified angle range. The discriminative module comprises a similarity discriminator to distinguish between authentic and synthetic images to ensure the image quality, and an angle discriminator to verify that generated images conform to the specified range of the azimuth angle. Combining these modules, the proposed methodology is capable of generating azimuth angle-controllable target images using only a limited number of support samples. The effectiveness of the proposed method is not only verified through various quality metrics, but also examined through the enhanced distinguishability of target recognition methods. In our experiments, we achieved SAR image generation within a given angle range on two datasets. In terms of generated image quality, our method has significant advantages over other methods in metrics such as FID and SSIM. Specifically, the FID was reduced by up to 0.37, and the SSIM was increased by up to 0.46. In the target recognition experiments, after augmenting the data, the accuracy improved by 6.16% and 3.29% under two different pitch angles, respectively. This demonstrates that our method has great advantages in the SAR image generation task, and the research content is of great value.

Large-scale, diverse, and high-quality data are the basis and key to achieving a good generalization of target detection and recognition algorithms based on deep learning. However, the existing methods for the intelligent augmentation of synthetic aperture radar (SAR) images are confronted with several issues, including training instability, inferior image quality, lack of physical interpretability, etc. To solve the above problems, this paper proposes a feature-level SAR target-data augmentation method. First, an enhanced capsule neural network (CapsNet) is proposed and employed for feature extraction, decoupling the attribute information of input data. Moreover, an attention mechanism-based attribute decoupling framework is used, which is beneficial for achieving a more effective representation of features. After that, the decoupled attribute feature, including amplitude, elevation angle, azimuth angle, and shape, can be perturbed to increase the diversity of features. On this basis, the augmentation of SAR target images is realized by reconstructing the perturbed features. In contrast to the augmentation methods using random noise as input, the proposed method realizes the mapping from the input of known distribution to the change in unknown distribution. This mapping method reduces the correlation distance between the input signal and the augmented data, therefore diminishing the demand for training data. In addition, we combine pixel loss and perceptual loss in the reconstruction process, which improves the quality of the augmented SAR data. The evaluation of the real and augmented images is conducted using four assessment metrics. The images generated by this method achieve a peak signal-to-noise ratio (PSNR) of 21.6845, radiometric resolution (RL) of 3.7114, and dynamic range (DR) of 24.0654. The experimental results demonstrate the superior performance of the proposed method.

Superpixel segmentation for polarimetric synthetic aperture radar (PolSAR) images plays a key role in remote-sensing tasks, such as ship detection and land-cover classification. However, the existing methods cannot directly generate multi-scale superpixels in a hierarchical style and they will take a long time when multi-scale segmentation is executed separately. In this article, we propose an effective and accurate hierarchical superpixel segmentation method, by introducing a minimum spanning tree (MST) algorithm called the Boruvka algorithm. To accurately measure the difference between neighboring pixels, we obtain the scattering mechanism information derived from the model-based refined 5-component decomposition (RFCD) and construct a comprehensive dissimilarity measure. In addition, the edge strength map and homogeneity measurement are considered to make use of the structural and spatial distribution information in the PolSAR image. On this basis, we can generate superpixels using the distance metric along with the MST framework. The proposed method can maintain good segmentation accuracy at multiple scales, and it generates superpixels in real time. According to the experimental results on the ESAR and AIRSAR datasets, our method is faster than the current state-of-the-art algorithms and preserves somewhat more image details in different segmentation scales.