As the infrared dim target has low signal-to-noise ratio and is difficult to be detected under complex background,an infrared dim target detection method based on partial differential equation is proposed.Firstly,the Gaussian template is used to smooth the image,and the impact of the strong edge is weakened.Then,the background is predicted through partial differential equation to get the residual image,and the segmentation threshold is determined based on the pixel number of statistical histogram to achieve target separation.Lastly,false alarms are rejected after multi-frame confirmation and target trajectory is exported.The results show that the two-dimensional entropy of the background is successful dropped to quarter of original background,and the detection probability of the target keeps above 94.6%.