The traditional detection algorithms have poor adaptability and weak anti-interference ability, which are vulnerable to the interference of the noise of clouds or mutation in dim target detection, and may lead to inaccurate detection results. To solve the problem, this paper puts forward an algorithm for dim target detection in the air based on information fusion. First of all, the image down-sampling is carried out to reduce the processing data, and the image is pre-processed by two kinds of filtering methods. Then, the traditional centroid detection and saliency detection are fused by model superposition. Finally, with the detection method based on the inter-frame information correlation in this paper, the “target-like” point is eliminated, so as to further improve the accuracy of the detection algorithm for small targets, and enhance the anti-interference ability of the detection algorithm. The algorithm is verified by simulations and contrast experiments. The experimental results show that the proposed algorithm has better detection effects and a stronger anti-interference ability.