A Background-Assisted Attention Detection Network for UAVs Detection in Pulse-Doppler Radar

Radar serves as an effective tool for detecting unmanned aerial vehicles (UAVs) and range-Doppler(R-D) map are frequently utilized to identify targets. Traditional methods and deep learning-based approaches often suffer from degraded detection performance in low signal-to-clutter ratio (SCR) environments. To address this issue, we propose the BAA-DetNet that leverages background information. Specifically, we introduce a background feature extractor(BFE) module designed to capture background features from the R-D map, along with a background-assisted attention (BAA) module aimed at suppressing background interference at the feature dimension, thereby enhancing detection performance in low SCR scenarios. Experimental results on both simulated and real datasets validate the effectiveness of our proposed model.