A ViT Merged Oriented-Detector with Neuron Attention for Ship Detection in SAR Images
Yiyang Huang, Di Wang, Wentao Huang, Daoxiang An · 2024
Due to the unique microwave properties of Synthetic Aperture Radar (SAR) images and the irregular distribution of vessels, deep learning techniques face significant challenges in the detection of ship targets within SAR imagery. This paper proposes an oriented ship target detection model based on the YOLOv8 algorithm, Neural Swin Transformer-YOLO (NST-YOLO). Experiments conducted on the Rotated Ship Detection Dataset in SAR Images (RSDD-SAR) and the SAR Ship Detection Dataset (SSDD+) and comparisons with the other oriented detection models demonstrate that the proposed NST-YOLO achieves state-of-the-art detection performance.