Aiming at the problems of high false alarm rate and low detection rate of traditional background suppression methods, an infrared small target image processing method based on improved filter and image weighted local entropy is proposed. The original infrared image is preprocessed by the improved median filter.The multi-scale gray difference operator and the local image entropy operator are constructed respectively, and the weighted local entropy is obtained by the dot product of the 2 operators, which can effectively suppress the background and eliminate the noise. The experimental results show that the method has good adaptability, and the operation efficiency is improved by 20.6% compared with the traditional median filtering algorithm, which has certain theoretical and engineering application value.