Infrared dim and small target detection based on three-dimensional collaborative filtering and spatial inversion modeling

摘要

The dim and small target detection in complex background is a challenging problem. Conventional methods may lead to high false alarm rates when targets are extremely dim and almost approximate to background noise. In this paper, a novel method named three-dimensional collaborative filtering and spatial inversion (3DCFSI) is proposed to deal with this problem. In order to suppress complex background clutter, a method based on three-dimensional cooperative filtering and differential calculation is first adopted. Then, a method based on energy accumulation is designed to enhance the dim target signal and improve the SNR. Finally, according to the constant false-alarm ratio judging and statistical method, target detection is realized by a spatial inversion model based on motion parameters. Experiments on real scenes demonstrate that the proposed method is able to stably detect small targets under different cloud clutter and background noise. Meanwhile, the proposed method has achieved more competitive performance than baseline methods.