Abstract：This paper presents an algorithm that can estimate dim and small target background via bilateral filtering. The conventional method directly uses bilateral filtering for background suppression, which can largely retain background, but it still maintains the small target so that the SCR of the subtracted image is not great, which brings difficulties to the later segmentation. The concept of the proposed filter is to utilize the new weight matrix having the structure reducing effects of the target pixels in order to predict exactly the background and exhibit higher contrast radio and SCR. The experimental results show that the proposed method has a better performance in comparison to the conventional one and the TDLMS filter.