Novel Infrared and Visible Images Fusion Algorithm Based on NSCT

Abstract

A novel fusion algorithm based on Nonsubsampled Contourlet Transform (NSCT) is proposed, according to the characteristics of infrared and visible images. Firstly, the registered infrared and visible images from the same scene were transformed by NSCT transforms; then the low frequency coefficient is fused by the combination of local energy and normalised correlation matrix, the high frequency coefficient fusion is fused by regional energy matching with regional variance; finally, the target image is obtained by performing inverse NSCT transforms. experimental results indicate that the proposed algorithm can effectively get more detail information and the fusion performance is dramatically better than traditional fusion methods.