Scene-Adaptive Infrared Small Target Detection Method

Infrared small target detection plays an important role in the tasks of public safety and area monitoring. However, there are often few target pixels in the image output by the system and the existing algorithm mainly accommodate the following shortcomings: (1) Insufficient robustness of the complex scenes. (2) Noise causes higher false alarms. To solve these problems, a scene-adaptive infrared small target detection method based on low-rank matrix recovery is proposed in this paper. Firstly, the infrared block image model is proposed to suppress the background and highlight the target. Secondly, adaptive parameters tuning model is designed to improve the robustness of the algorithm under complex background with various noises. Finally, the Kalman filter parameter estimation and update strategy using multi-frame information is established to suppress background clutter false alarm and enhance target detection effect. Compared with the existing infrared target detection methods, the proposed algorithm is more robust and has better detection effect.