Ship detection from synthetic aperture radar (SAR) remote sensing images is essential for monitoring water traffic and marine safety. Numerous methods for ship detection have been developed; however, the detection of small ships presents unique challenges. SAR image characteristics, such as the sidelobe effect and blurred outline induced by the special imaging mechanism, as well as the small ship size, are the primary factors that lower the detection accuracy. This paper provides a sidelobe-aware small ship detection network for synthetic aperture radar imagery. First, considering the sidelobe effect and blurred outline, dual-pooling, i.e., average pooling and max pooling, was utilized to build a feature extraction module that lowered the effects of strong scattering points outside of the ship body and enhanced the ship body information. Second, as the bipartition process of the average pooling and maximum pooling caused some loss of original data information, different feature maps in the network were concatenated to construct a new network structure to compensate for the information lost and enrich the small ship features. Third, because the traditional loss function based on centroid distance and aspect ratio may result in the same loss function value for different prediction box sizes, a novel loss function based on the dual Euclidean distances of the corner point coordinates between the prediction box and the real box was proposed, which could accurately describe various overlapping box situations. Experiments using the Large-Scale SAR Ship Detection Dataset (LS-SSDD), SAR Ship Detection Dataset (SSDD), and AIR-SARShip dataset validated the efficacy and state-of-the-art performance.

In existing superpixel-wise segmentation algorithms, superpixel generation most often is an isolated preprocessing step. The segmentation performance is determined to a certain extent by the accuracy of superpixels. However, it is still a challenge to develop a stable superpixel generation method. In this article, we attempt to incorporate the superpixel generation and merging steps into an end-to-end trainable deep network. First, we employ a recently proposed differentiable superpixel generation method to over-segment the single-polarization synthetic aperture radar (SAR) image. It outputs the statistical likelihood that each pixel belongs to different superpixels. In superpixel merging part, as one of our main contributions, we propose a superpixel-wise statistical dissimilarity measure method for converting the soft superpixels set into a self-connected weighted graph. More importantly, inspired by the concept of the number of walks in graph theory, we define the $k$ -order connectivity of each vertex. This definition can intelligently indicate the potential soft cluster centers and class assignments in graph. This merging method is differentiable, computationally simple, and free of empirical parameters. The superpixel generation and merging phases can be implemented under a unified deep network. The benefit is that our method can iteratively adjust the shapes of the superpixels according to the boundaries and segmentation results during training, until the satisfactory segmentation results are captured. Experimental results on real SAR images demonstrate that the segmentation precision of our proposed method is superior to other state-of-the-art methods in terms of precision and computational efficiency.

The recent emergence of high-resolution Synthetic Aperture Radar (SAR) images leads to massive amounts of data. In order to segment these big remotely sensed data in an acceptable time frame, more and more segmentation algorithms based on deep learning attempt to take superpixels as processing units. However, the over-segmented images become non-Euclidean structure data that traditional deep Convolutional Neural Networks (CNN) cannot directly process. Here, we propose a novel Attention Graph Convolution Network (AGCN) to perform superpixel-wise segmentation in big SAR imagery data. AGCN consists of an attention mechanism layer and Graph Convolution Networks (GCN). GCN can operate on graph-structure data by generalizing convolutions to the graph domain and have been successfully applied in tasks such as node classification. The attention mechanism layer is introduced to guide the graph convolution layers to focus on the most relevant nodes in order to make decisions by specifying different coefficients to different nodes in a neighbourhood. The attention layer is located before the convolution layers, and noisy information from the neighbouring nodes has less negative influence on the attention coefficients. Quantified experiments on two airborne SAR image datasets prove that the proposed method outperforms the other state-of-the-art segmentation approaches. Its computation time is also far less than the current mainstream pixel-level semantic segmentation networks.

Since the number of superpixels is lower than that of pixels, superpixels can substantially speed up subsequent processing steps and have been widely used in synthetic aperture radar (SAR) image segmentation. However, in most of the existing superpixel-wise segmentation algorithms, superpixel prediction is an isolated preprocessing step and is independent of the segmentation task. The performance of the segmentation results is determined by the accuracy of superpixels. Once superpixels are generated, their shape cannot be changed in the following segmentation stage, even if the same superpixels contain pixels of different landcovers. To address this, we propose an end-to-end trainable superpixel-wise segmentation method for single-polarization SAR images. First, we design a differentiable boundary-ware clustering method for estimating task-specific superpixels. Instead of the hard association between pixels and superpixels in the existing superpixel algorithms, this method introduces the soft association map to make the clustering differentiable. Hence, it can be implemented using a simple deep fully convolutional network. In the segmentation part, we propose a novel soft graph convolution network (Soft-GCN), which takes the association map as input and performs superpixel-wise segmentation. The advantage of our method is that superpixel generation and graph convolution parts can be trained under a unified framework, until two parts obtain the optimum parameters. In the training process, it can adaptively adjust the shape of the superpixels according to the segmentation results, ensuring the superpixels correctly adhere the boundaries. Experimental results with simulated and real SAR images demonstrate that our method outperforms other state-of-the-art segmentation algorithms, while also being faster.

In existing superpixel-wise saliency detection algorithms, superpixel generation often is an isolated preprocessing step. The performance of saliency maps is determined by the accuracy of superpixels to a certain extent. However, it is still a challenge to develop a stable superpixel generation method. In this paper, we attempt to incorporate the superpixel generation and saliency calculation steps into an end-to-end trainable deep network. First, we employ a recently proposed differentiable superpixel generation method to over-segment the SAR images, which outputs the possibility that the pixels assigned to neighbor superpixels (soft superpixel). In saliency calculation part, as one of our main contributions, we propose a differentiable and computationally simple saliency model, i.e., lacunarity cue. It is inspired by the fact that generally the backscattering intensity of regions of interest (ROIs) in SAR images irregularly fluctuates. While the areas with consistent pixels are often ignored as the clusters. We improve the pixel-wise box differential dimension algorithm to measure the irregularity of scattering points in a superpixel. The superpixel generation and saliency calculation can be implemented under a unified deep network. Hence the shapes of the superpixels can be iteratively adjusted according to the saliency maps until the ROIs are correctly detected. Experiments on real SAR images with different sizes and scenes show that the saliency maps can effectively highlight the target areas, outperforming the state-of-the-art saliency detection models.

Accurate detection of rivers plays a significant role in water conservancy construction and ecological protection, where airborne Synthetic Aperture Radar (SAR) data has already become one of the main sources. However, extracting river information from radar data efficiently and accurately still remains an open problem. The existing methods for detecting rivers are typically based on rivers’ edges, which are easily mixed with those of artificial buildings or farmland. In addition, pixel based image processing approaches cannot meet the requirement of real time processing. Inspired by the feature integration and target recognition capabilities of biological vision systems, in this paper, we present a hierarchical method for automated detection of river networks in the high-resolution SAR data using biologically visual saliency modeling. For effective saliency detection, the original image is first over-segmented into a set of primitive superpixels. A visual feature (VF) set is designed to extract a regional feature histogram, which is then quantized based on the optimal parameters learned from the labeled SAR images. Afterwards, three saliency measurements based on the specificity of the rivers in the SAR images are proposed to generate a single layer saliency map, i.e., Local Region Contrast (LRC), Boundary Connectivity (BC) and Edge Density (ED). Finally, by exploiting belief propagation, we propose a multi-layer saliency fusion approach to derive a high-quality saliency map. Extensive experimental results on three airborne SAR image datasets with the ground truth demonstrate that the proposed saliency model consistently outperforms the existing saliency target detection models.