基于非降采样Contourlet变换（NSCT）的红外与可见光图像融合及稳健性分析

提出一种基于区域显著性融合规则的非降采样Contourlet 变换的红外与可见光图像融合方法. 首先, 对来自同一场景图像的红外与可见光图像的??分量进行非降采样Contourlet变换; 然后, 依照区域匹配度量和显著性度量规则进行融合, 而融合图像通过非降采样Contourlet反变换即可得到. 最后, 针对此方法进行大量实验, 并将其融合结果与基于小波变换及拉普拉斯金子塔变换的融合结果进行比较, 同时分析它们在不同噪声条件下的性能指标.

Infrared and Visible Light Images Fusion Method Based on NSCT Its Robustness Analysis

An infrared and visible image fusion method based on the non-subsampled Contourlet transform(NSCT) is
proposed. With two distinct modes of fusion method in low-frequency and in high-frequency separately, the fused image is obtained by inverse NSCT. Compared with the fusion method based on wavelet transform and Laplace-pyramid transform under different noise conditions, the experiments show that this approach can achieve good results.