In the current single-frame image threshold segmentation, the segmentation is susceptible to mutations, the background segmentation is not obvious and the segmentation effectiveness is poor. To solve these problems, an automatic threshold segmentation method based on infrared image frame association is proposed. Firstly, the automatic threshold segmentation method is used to segment a single-frame image simply. Secondly, the image is processed group by group according to the frame-related information. Then, the weights are assigned to each frame. Finally, the segmentation threshold of each frame is determined to improve the anti-interference and segmentation effectiveness. The validity and feasibility of the method are verified through theoretical analysis and experimental simulation. Compared with other algorithms, the proposed method is stronger in anti-interference ability. It can segment the target image clearly from the background and has better segmentation effectiveness and applicability.