Decision Fusion Moving Target Detection of Radar Video Based on D-S Evidence Theory

In dense clutter and complex multi-objective situation, radar automatic detection and tracking will produce many false tracks. In order to solve this problem, first, using iterative threshold segmentation method to rapidly detect the suspected targets from a single frame radar video image, and gaining the continuous multiframe detection results by accumulation. Then, synthesize single frame detection results and multiple frames accumulation results to extract the area change ratio and energy density of targets, and build their basic probability assignment (BPA) functions, respectively. In addition, D-S evidence theory is adopted to carry out decision fusion, to obtain the optimal decisions in order to realize the automatic detection of moving targets. Finally, the effectiveness of the algorithm is verified by experimental results.