Abstract: Aiming at the problems of low detection accuracy and high false alarm rate of infrared weak and small targets due to violent background changes, more clutter and low signal-to-noise ratio in complex scenes, an infrared weak target detection algorithm combining image global information and local contrast is proposed. The algorithm uses the variance and signal-to-noise ratio to statistically analyze all pixels of the image, and processes the image globally to obtain a feature map, so as to adapt to the complex background with sharp gray level changes, while suppressing a large number of flat background clutter and improving the signal-to-noise ratio of the target. Aiming at the strong background edge noise and bright pixel noise mainly existing in the feature map, the weighted absolute directional mean difference (WADMD) algorithm is used to calculate the absolute average difference between the target and the background as the weighting coefficient, and the judgment threshold is used to suppress the negative contrast, and suppress the high luminance noise, and improve the significance of the target. Experiments show that compared with the comparison algorithm, the proposed algorithm can adapt to the changeable complex background, and improve the signal-to-noise ratio of the target more obviously, and have better robustness.