To detect and track targets accurately when attacking small targets, while attempting to achieve a low signal-to-noise ratio and few pixels of small and dim targets in the air, images are preprocessed based on infrared video images using Gauss filtering and the top-hat operator in this study. The edge detection algorithm is used to detect and locate the target in the image. The initial position of the target is continuously tracked by the kernel correlation filter tracking algorithm. Finally, the tracking effect is quantitatively evaluated. Experimental results show that the tracking error of the maximum field of view angle does not exceed 0.0062., and that the average running speed can reach 25.3 fps/s. This method can effectively detect and track small and dim infrared targets in the air.