Intelligent Target Detection for Airborne STAP Radar Based on Space-Time Image Feature and YOLO

For airborne space-time adaptive processing (STAP) radar, this paper proposes an intelligent method for detecting slow-moving targets without estimating the clutter covariance matrix. The proposed method uses the sparse Bayesian learning (SBL) algorithm to generate a high-quality space-time image of clutter and target. Then, the image feature difference of clutter and target is analyzed. Finally, the you-only-look-once (YOLO) deep neural network (DNN)-based detector is used to discriminate the target from the clutter for detection purposes. Simulation results obtained under different conditions show that the proposed algorithm can achieve high detection performance of slow-moving targets.