Aiming at the difficulty of target recognition in infrared images, a method for identifying weak and small moving targets in infrared images based on background difference method is proposed. Firstly, a spatiotemporal regularization correlation filtering algorithm is used to process a large range of background areas near the target, and side window filtering is used to further remove noise, denoise and preserve edges in low-quality infrared images, and improve image quality. Then, the background difference method is used to preliminarily identify weak and small targets in the image, Finally, based on the preliminary recognition results, Kalman filtering is introduced to predict the motion trajectory position of weak targets in infrared images, achieving weak motion target recognition. The experimental results show that the proposed method has a higher recognition success rate and more accurate recognition position.