Research on the infrared detection method of anti “low slow small” UAVs

DOI:	10.11809/bqzbgcxb2023.07.027
Keywords:	anti UAVs; density distance space; single frame detection; small infrared targets; equivalent localization
Abstract:	In view of the aerial threat of “low slow small” UAVs, all countries have vigorously invested in anti UAVs technology research. This paper presents a new method of infrared single frame detection based on density distance space. The concept of density and distance is proposed and the density distance space is established based on the fact that the small infrared target has a higher gray value in a limited area than in the adjacent area. The irregular target regions are converted into regular geometric figures to obtain the accurate target positions through the equivalent target localization method. The experimental results show that the density distance detection algorithm has better detection performance and robustness under complex backgrounds. When the target is obvious and there is continuous background clutter, the AUC value is close to 1. The average detection time per frame is 0.021 2 s. In addition, it takes into account the detection accuracy and detection speed. The algorithm is not affected by the size and shape changes of a UAV, and has strong adaptability, which is suitable for small single frame infrared target detection under various backgrounds.
Issue:	Vol. 44 No. 7 (2023)
Published:	2023-07-28
PDF HTML