Background suppression is a key technique to the automatic detection and tracking small and dim objects.And adaptive background estimation is an efficient way to realize this technique.To the problem.an M-filter adaptive background suppression algorithm is presented.In the algorithm,the image background is considered as the mixed interference by the target and observing noises,and the true image background is estimated adaptively via M-estimation.This algorithm is robustness and can resist high mixed interference.Simulations and experiments show that the algorithm is more efficient to estimate the background and improve signal to noise ratio(SNR) im mixed noises compared with the mean filter method,the median filter method and the Wiener filter method.