A Self-Attention Dictionary Learning-Based Method for Ship Detection in SAR Images

Data-driven algorithms based on deep neural networks (DNNs) for ship detection in synthetic aperture radar (SAR) images are restricted by limited training samples and complex background interference. Inspired by the sparsity and neighborhood relevance of ships in SAR images, a novel detection algorithm based on self-attention dictionary learning (SADL) is proposed in this letter, which only requires a few samples for training. A self-attention mechanism is injected to learn discriminative features between classes, which can extract inherent information from sequences adaptively. Particularly, a hybrid loss function is tailored for the representations of multiclasses targets using SADL, which consists of the reconstruction error, minimal intraclass error, maximum interclass error, and exclusiveness error. Further, a SADL-based ship detection method is proposed by building subdictionaries of the target and background, respectively. The gradient and intensity information in a fixed neighborhood are used to construct the feature dictionary to suppress complex background interference and provide effective prior knowledge. Experiments conducted on the large-scale SAR ship detection dataset (LS-SSDD-v1.0) demonstrate the effectiveness of the proposed method, which achieves an F1-score of 0.47 on 3000 test images with only three training images.