Abstract: Unmanned aerial vehicle (UAV) target detection is the key to low-altitude security. To address detail loss, insignificant targets, and poor visual effects in fusing infrared intensity and polarization images under complex backgrounds, this study proposes a joint fusion algorithm using the stationary wavelet transform (SWT) and non-subsampled contourlet transform (NSCT) to enhance target saliency while ensuring fusion quality. SWT first decomposes the source image. The low-frequency images were subjected to further NSCT decomposition. Adaptively weighted fusion was applied to the NSCT low-frequency sub-images, whereas high-frequency sub-images were fused using the maximum regional edge density and then reconstructed via the NSCT inverse transform. Next, intralayer contrast fusion was applied to the high-frequency SWT images. Finally, the fused images were subjected to SWT reconstruction and enhancement to generate the final results. Comparative experiments with seven classic algorithms on six image sets and stability tests on 45 image sets from multiple scenes showed that the proposed method outperformed single algorithms in terms of visual effects and feature preservation, exhibited better robustness, significantly improved scene understanding for computer vision, and enhanced UAV target detection and recognition.