Effective visualization of infrared images with low contrast and low signal-to-noise ratio is one of the key technologies for high-performance infrared imaging systems. The conventional decomposition-based algorithms have advantages in image details enhancement, but still suffer from high computational cost, unbalanced noise suppression and detail information. In this paper, we decompose the base and detail components of the image by the iterative least squares and the Difference of Gaussian filter, and further enhance the base layer and the detail layer via plateau equalization and gradient mask, respectively. We then fusion the enhancement result and re-project to 8-bit dynamic range. Experimental results shown that the proposed method achieves a good balance between detail enhancement and computational cost, with a high-performance in different scenes. © 2023 SPIE.