TSDet: End-to-End Method with Transformer for SAR Ship Detection

In recent years, the growing demand for synthetic aperture radar (SAR) ship detection on mobile and satellite platforms has made lightweight detection models an important research focus. However, existing SAR ship detection methods are often limited by high computational cost and inefficient utilization of intrinsic SAR characteristics, particularly the loss of edge and high-frequency information during downsampling and the insufficient interaction of multi-scale features across network layers. To address these issues, we propose a lightweight and efficient detection framework named ESENet, which is designed from the perspective of preserving discriminative information, improving cross-scale feature interaction, and reducing structural redundancy. Specifically, a lightweight feature extraction module is introduced to alleviate high-frequency and edge information loss caused by conventional pooling operations, thereby enhancing the representation of small and low-contrast ship targets commonly present in SAR images. Furthermore, a Slender bidirectional feature pyramid network is designed by streamlining cross-layer connections and reducing redundant fusion paths, enabling effective multi-scale feature aggregation with lower computational complexity while strengthening edge-aware feature propagation. In addition, an optimized detection head with parameter sharing across scales is employed to further reduce model parameters and computational overhead without sacrificing detection accuracy or inference speed. Extensive experiments conducted on multiple public SAR ship detection datasets demonstrate the effectiveness of the proposed ESENet, achieving average accuracy rates of 91.1% on HRSID, 98.9% on SSDD, 97.3% on SSDD+, and 72.1% on synthetic and real-world small-object ship detection dataset, while maintaining a compact model size of only 1.3 M parameters, which confirms its suitability for resource-constrained SAR ship detection applications.

In light of recent advances in deep learning and Synthetic Aperture Radar (SAR) technology, there has been a growing adoption of ship detection models that are based on deep learning methodologies. However, the efficiency of SAR ship detection models is significantly impacted by complex backgrounds, noise, and multi-scale ships (the number of pixels occupied by ships in SAR images varies significantly). To address the aforementioned issues, this research proposes a Pyramid Pooling Attention Network (PPA-Net) for SAR multi-scale ship detection. Firstly, a Pyramid Pooled Attention Module (PPAM) is designed to alleviate the influence of background noise on ship detection while its parallel component favors the processing of multiple ship sizes. Different from the previous attention module, the PPAM module can better suppress the background noise in SAR images because it considers the saliency of ships in SAR images. Secondly, an Adaptive Feature Balancing Module (AFBM) is developed, which can automatically balance the conflict between ship semantic information and location information. Finally, the detection capabilities of the ship detection model for multi-scale ships are further improved by introducing the Atrous Spatial Pyramid Pooling (ASPP) module. This innovative module enhances the detection model’s ability to detect ships of varying scales by extracting features from multiple scales using atrous convolutions and spatial pyramid pooling. PPA-Net achieved detection accuracies of 95.19% and 89.27% on the High-Resolution SAR Images Dataset (HRSID) and the SAR Ship Detection Dataset (SSDD), respectively. The experimental results demonstrate that PPA-Net outperforms other ship detection models.

Synthetic Aperture Radar (SAR) Ship Detection (SSD) is an important application, and it has been widely used in commercial and military fields. With the development of remote sensing technology, the use of a large amount of increasing unlabeled data is a challenge for model training, and the interference caused by the reefs and the nearshore facilities is also a challenge for SAR ship detection. In this work, we propose a Feature Integration Attention Detection (FIAD) Network to detect ships in SAR images, namely the FIAD Net. The FIAD Net has two contributions: a ModiBYOL method based on self-supervised learning and a novel Feature Integration Attention (FIA) module. The FIA module could enhance the feature learning ability of the backbone network, and the ModiBYOL could utilize the unlabeled data as a source for backbone pre-training. Experiments on the OpenSARShip dataset, the SAR ship detection dataset (SSDD), and the SAR Ship dataset show that the proposed method achieves an accuracy comparable to the state-of-art methods with a speed of 47 Frames Per Second (FPS). Besides, the effectiveness of the proposed modules is studied through an ablation study.

During forming synthetic aperture radar (SAR) image, the inherent speckle noise severely affects the application result of SAR images. To deal with this problem, based on the contour bougie morphology (CB morphology) and the generalized multiple structuring elements morphology, a novel series morphological filter was constructed to suppress the speckle noise in SAR images. Firstly, by use of the omnidirectional multiple structuring elements the generalized multi-scale morphology open-max filter was adopted to filter the noise at the cost of minimal loss of image geometrical details, and then a CB open and a CB close series filter were used to filter the remained speckle noise. Thus it can not only effectively suppress the speckle noise in images but also preserve more detailed information of the images. Lastly, the simulated SAR images and the actual SAR images are simulated, and the results prove that the algorithm presented in this paper can efficiently suppresses the speckle noise in SAR images.

Ship detection in synthetic aperture radar (SAR) images is becoming a research hotspot. In recent years, as the rise of artificial intelligence, deep learning has almost dominated SAR ship detection community for its higher accuracy, faster speed, less human intervention, etc. However, today, there is still a lack of a reliable deep learning SAR ship detection dataset that can meet the practical migration application of ship detection in large-scene space-borne SAR images. Thus, to solve this problem, this paper releases a Large-Scale SAR Ship Detection Dataset-v1.0 (LS-SSDD-v1.0) from Sentinel-1, for small ship detection under large-scale backgrounds. LS-SSDD-v1.0 contains 15 large-scale SAR images whose ground truths are correctly labeled by SAR experts by drawing support from the Automatic Identification System (AIS) and Google Earth. To facilitate network training, the large-scale images are directly cut into 9000 sub-images without bells and whistles, providing convenience for subsequent detection result presentation in large-scale SAR images. Notably, LS-SSDD-v1.0 has five advantages: (1) large-scale backgrounds, (2) small ship detection, (3) abundant pure backgrounds, (4) fully automatic detection flow, and (5) numerous and standardized research baselines. Last but not least, combined with the advantage of abundant pure backgrounds, we also propose a Pure Background Hybrid Training mechanism (PBHT-mechanism) to suppress false alarms of land in large-scale SAR images. Experimental results of ablation study can verify the effectiveness of the PBHT-mechanism. LS-SSDD-v1.0 can inspire related scholars to make extensive research into SAR ship detection methods with engineering application value, which is conducive to the progress of SAR intelligent interpretation technology.

In general, synthetic aperture radar (SAR) imaging and image processing are two sequential steps in SAR image processing. Due to the large size of SAR images, most image processing algorithms require image segmentation before processing. However, the existence of speckle noise in SAR images, as well as poor contrast and the uneven distribution of gray values in the same target, make SAR images difficult to segment. In order to facilitate the subsequent processing of SAR images, this paper proposes a new method that combines the back-projection algorithm (BPA) and a first-order gradient operator to enhance the edges of SAR images to overcome image segmentation problems. For complex-valued signals, the gradient operator was applied directly to the imaging process. The experimental results of simulated images and real images validate our proposed method. For the simulated scene, the supervised image segmentation evaluation indexes of our method have more than 1.18%, 11.2% and 11.72% improvement on probabilistic Rand index (PRI), variability index (VI), and global consistency error (GCE). The proposed imaging method will make SAR image segmentation and related applications easier.

Synthetic aperture radar (SAR) ship detection based on deep learning has been widely applied in recent years. However, two main obstacles are hindering SAR ship detection. First, the identification of ships in a port is seriously disrupted by the presence of onshore buildings. It is difficult for the existing detection algorithms to effectively distinguish the targets from such a complex background. Additionally, it appears more complicated to accurately locate densely arranged ships. Second, the ships in SAR images exist at a variety of scales due to multiresolution imaging modes and the variety of ship shapes; these pose a much greater challenge to ship detection. To solve the above problems, this paper proposes an object detection network combined with an attention mechanism to accurately locate targets in complex scenarios. To address the diverse scales of ship targets, we construct a loss function that incorporates the generalized intersection over union (GIoU) loss to reduce the scale sensitivity of the network. For the final processing of the results, soft nonmaximum suppression (Soft-NMS) is also introduced into the model to reduce the number of missed detections for ship targets in the presence of severe overlap. The experimental results reveal that the proposed model exhibits excellent performance on the extended SAR ship detection dataset (SSDD) while achieving real-time detection.

Speckle noise is a kind of disturbance which appears as granular noise in Synthetic Aperture Radar (SAR) images. It makes the images quite difficult to interpret. Over the years, many filters have been used to mitigate this noise, with limited success. This work will primarily compare the Mean, Median, Guided Image Filter and a proposed filter for their effectiveness in reducing the speckle noise in SAR images. To measure it quantitatively, standard metrics like Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM) are used. They help in assessing, and making a base to compare these filters. The comparison shows that the proposed filter and the Guided Image Filter, perform better in reducing noise and maintaining the structure of the image compared to older methods like Mean and Median Filters. This study may help in identification of new strategies for reducing speckle noise in SAR images, offering valuable insights for improving SAR image processing.

Synthetic aperture radar (SAR) image segmentation usually involves two crucial issues: suitable speckle noise removing technique and effective image segmentation methodology. Here, an efficient SAR image segmentation method considering both of the two aspects is presented. As for the first issue, the famous nonlocal mean (NLM) filter is introduced in this study to suppress the multiplicative speckle noise in SAR image. Furthermore, to achieve a higher denoising accuracy, the local neighboring pixels in the searching window are projected into a lower dimensional subspace by principal component analysis (PCA). Thus, the nonlocal mean filter is implemented in the subspace. Afterwards, a multi-objective clustering algorithm is proposed using the principals of artificial immune system (AIS) and kernel-induced distance measures. The multi-objective clustering has been shown to discover the data distribution with different characteristics and the kernel methods can improve its robustness to noise and outliers. Experiments demonstrate that the proposed method is able to partition the SAR image robustly and accurately than the conventional approaches.

Synthetic Aperture Radar (SAR) ship detection has been a research hotspot and is significant for marine surveillance. Traditional constant false alarm rate (CFAR) detector has the disadvantages of high false alarm and poor adaptability. Deep learning provides a unique solution for SAR ship detection. However, the traditional deep network cannot reach very deep thus the accuracy is limited, and the training speed is slow. In this paper, a very deep network ResNet with higher accuracy and faster training speed is applied to train the SAR ship detection model. Moreover, transfer learning is applied to combat the small dataset. The proposed method is tested on a general SAR ship dataset and achieves 94.7% average precision. Comparative experiments show that our method has the best performance and which verifies the effectiveness of our method.

Synthetic aperture radar (SAR) and optical sensing are two important means of Earth observation. SAR-to-optical image translation (S2OIT) can integrate the advantages of both and assist SAR image interpretation under all-day and all-weather conditions. The existing S2OIT methods generally follow a paired training paradigm, which is difficult when dealing with the unpaired S2OIT application scenarios. Moreover, the generator and discriminator in current S2OIT methods have insufficient scene memory for SAR images, resulting in regional landform deformation in the generated images. To address these issues, we propose a novel generative adversarial network capable of memorizing scene for unpaired S2OIT called MS-GAN. The cycle learning framework based on cycle generative adversarial network for unpaired S2OIT is designed to construct the translation mapping between unpaired SAR and optical images. The multiscale representation generator is constructed for multiscale fusion and utilization of scene features of SAR images. The proposed multireceptive field discriminator has the ability to enhance scene memory and generate higher quality optical images in different landforms. In addition, the designed subbands shrinkage denoising module can further suppress the effect of speckle noise in SAR images on the quality of the generated results. Extensive experiments conducted on three challenging datasets SEN1-2, WHU-SEN-City, and QXS-SAROPT demonstrate that the proposed MS-GAN outperforms the state-of-the-art methods on both subjective and objective evaluation metrics.

Due to its great application value in the military and civilian fields, ship detection in synthetic aperture radar (SAR) images has always attracted much attention. However, ship targets in High-Resolution (HR) SAR images show the significant characteristics of multi-scale, arbitrary directions and dense arrangement, posing enormous challenges to detect ships quickly and accurately. To address these issues above, a novel YOLO-based arbitrary-oriented SAR ship detector using bi-directional feature fusion and angular classification (BiFA-YOLO) is proposed in this article. First of all, a novel bi-directional feature fusion module (Bi-DFFM) tailored to SAR ship detection is applied to the YOLO framework. This module can efficiently aggregate multi-scale features through bi-directional (top-down and bottom-up) information interaction, which is helpful for detecting multi-scale ships. Secondly, to effectively detect arbitrary-oriented and densely arranged ships in HR SAR images, we add an angular classification structure to the head network. This structure is conducive to accurately obtaining ships’ angle information without the problem of boundary discontinuity and complicated parameter regression. Meanwhile, in BiFA-YOLO, a random rotation mosaic data augmentation method is employed to suppress the impact of angle imbalance. Compared with other conventional data augmentation methods, the proposed method can better improve detection performance of arbitrary-oriented ships. Finally, we conduct extensive experiments on the SAR ship detection dataset (SSDD) and large-scene HR SAR images from GF-3 satellite to verify our method. The proposed method can reach the detection performance with precision = 94.85%, recall = 93.97%, average precision = 93.90%, and F1-score = 0.9441 on SSDD. The detection speed of our method is approximately 13.3 ms per 512 × 512 image. In addition, comparison experiments with other deep learning-based methods and verification experiments on large-scene HR SAR images demonstrate that our method shows strong robustness and adaptability.

Due to the extensive application and good performance of the synthetic aperture radar(SAR), the study to SAR image becomes more and more important. But the intensive analysis to the SAR image is hampered by the speckle noise within it. Based on the knowledge of SAR imaging system, an improved method that uses the adaptive window to suppress the speckle noise in SAR images is proposed in this paper. Compared with the classical fix\|window spatial filtering method, this proposed one can adjust the size of filtering window adaptively according to the statistical of pixels within the window, and it can well suppress the speckle noise in the homogenous area and preserve the detail information in heterogenous area. In simulating experiment, this new method is proved good performance in reducing speckle and preserving edges at the same time.

A growing number of researchers begin to use the arbitrary-oriented detector to detect ships. Because in the remote sensing dataset, ships are shown to be narrow and dense. Most of the deep learning methods used in synthetic aperture radar (SAR) ship detection are the same as or variants of those of optical remote sensing. However, due to the relatively low resolution and signal-to-noise ratio, the amplitude information in SAR image becomes more contaminated than that in optical image. Therefore, it is very difficult to train the arbitrary-oriented target detection task based on SAR dataset. In order to solve this problem, we attempt to further extract the frequency domain features of the SAR image by introducing a frequency attention module. The discrete cosine transform (DCT) to converts SAR image from spatial domain to frequency domain. Experimental results on HRSID datasets show that the proposed method can significantly improve the performance of arbitrary-oriented SAR ship detection.

Synthetic Aperture Radar (SAR) images are strongly corrupted by the speckle noise due to random electromagnetic waves interference. The speckle noise reduces the quality of images and makes their interpretation and analysis really difficult, so it's necessary to filter images to remove the noise in order to preserve as much as possible the most important features of the signal. To achieve this goal, in this paper we present an efficient method that reduces the speckle noise in SAR images, based on the Contourlet Transform (CT). The CT is a new image decomposition scheme that provides sparse representation of the data, constructed by combining two successive stages, applying in first a Laplacian pyramidal decomposition followed by a directional filter bank. This non-linear approach is designed to give a good representation of the geometrical content of the image. Recently, the Stationary version of the Contourlet Transform (SCT) has been proposed to preserve the shift-invariant property. In the present paper, we explore two different de-noising methods: the Bayesian Shrinkage based on a weighting factor that reduces noise by using the contourlet coefficients, and the Soft Thresholding based on the choice of the threshold that ensures adaptation to the noiseless signals. Hence, we present a comparative study of the results obtained through the SCT considering different stages of decomposition's levels and different kind of filters, and the Lee Adaptive Filter. A performance evaluation is realized to validate our methods.