In view of the interference of ghosting phenomenon to target detection, which is caused by the observation equipment in range experiment, and the interference of clouds to dim small target in the sky background, an algorithm for aerial infrared small target detection is proposed based on partial row pixel contrast. The algorithm can be used for detecting aerial infrared dim targets under low SNR condition. The method uses median filter and morphological filtering to obtain filtered image, extracts the region of interest from the filtered image based on the pixel feature of the target area;then uses the partial row pixel contrast to detect the position of the target. The method can greatly improve the detection efficiency with less time. This paper introduces and analyzes the theoretical effectiveness of the infrared small target image detection algorithm, and makes comparison of it with other algorithms through the Matlab simulation. The experimental results show that the proposed algorithm can detect the aerial infrared dim small targets more effectively under the condition of low SNR with higher accuracy and better performance.