SAR Aircraft Detection based on Scattering Enhancement and Location Attention network

Due to the interference caused by the complex background and the discreteness of aircraft scattering, the task of detecting aircraft in synthetic aperture radar (SAR) images remains a challenging one. To address these problems, a novel method named scattering enhancement and location attention network (SELANet) is here proposed to detect aircraft via combining traditional image processing and deep learning together. At first, a scattering information enhancement module (SIEM) is proposed to highlight the scattering points of aircraft targets. Moreover, to more effectively focus on the location of aircraft targets, a spatial location attention module (SLAM) is further designed. Experiments are performed on the SADD dataset and the results indicate that SELANet is more apt for aircraft detection. On average, the Precision, Recall, and F1-score values are respectively 3.6%, 2.9%, and 3.2% higher than the baseline of standard YOLOv8s algorithm.