Detection and Tracking of Infrared Small Targets in Complex Dynamic Backgrounds by Using Image Registration and Pipeline Filter

Abstract

This paper propses an infrared small target detection method in complex dynamic backgrounds in sequence images by using image registration and pipeline filter. For dynamic scenes caused by camera movements, registration of image frames and reference frame images is conducted by using the feature point matching algorithm. Next, the frame difference method is used to detect small mobile target in video image sequences. Finally, the impact of ground clutter noise and random noise are eliminated by using a pipeline filter to precisely track the trajectory of small moving targets. The experiment results show that the proposed method is superior to the classic infrared small target detection algorithm in terms of accuracy and stability in complex ground backgrounds.