New Adaptive Background Suppression Algorithm via One Step M-filter

Abstract

Background suppression is a key technique to small and dim targets automatic detection and tracking. And adaptive background estimation is an efficient way to realize this technique. In this paper, the one step M-filter (OSMF) adaptive background suppression algorithm is presented. In the algorithm, targets and observing noises are considered as mixed noise interference of the image background and then the true image background is estimated adaptively via M-estimation. Firstly, initial value for background clutters is estimated applying the order statistics filter. Then the final estimation of background clutters is given using the OSMF. Finally, background suppression is done by removing background clutters from original images. The proposed method is applied to simulations and real visible image data to demonstrate the robustness and effectiveness of the OSMF