为使红外图像与可见光图像融合后的图像能获得更多目标信息和细节信息，本文提出了一种基于显著性图的图像融合方法。使用改进的 Frequency Tuned(FT)算法提取红外图像的显著性图，并使用对比度受限的自适应直方图均衡化(Contrast Limited Adaptive Histgram Equalization，CLAHE)算法增强可见光图像的对比度。将红外图像与增强后的可见光图像进行非下采样轮廓波变换(Nonsubsampled Contourlet Transform，NSCT)后，根据所设定的融合规则分别对红外与可见光图像的低频部分与高频部分进行融合，最后对融合系数进行 NSCT逆变换操作后获到融合图像。实验表明，该融合方法相较于其他方法而言，保留了更多的目标信息和细节信息，可以取得更好的视觉效果。

To obtain more details and target information from fused infrared and visible images, an image fusion method based on saliency map is proposed. First, the frequency-tuned algorithm is improved to extract the saliency map of the infrared image, and the contrast of visible image is enhanced by contrast limited adaptive histogram equalization. Then, the infrared and enhanced visible images are decomposed by nonsubsampled contourlet transform, and the low and high frequency coefficients are fused according to the fusion rules. Finally, the fusion image is obtained by the inverse transformation of the low and high frequency fusion coefficients. Experimental results show that compared to the other methods, this fusion method retains more details and target information with good visual effects.