Fusion of infrared and visible images based on nonsubsampled contourlet transform and sparse K-SVD dictionary learning

In this paper, an image fusion method, which is named NSCT_SK_SVD, is proposed for infrared and visible images, where Nonsubsampled Contourlet Transform (NSCT) and sparse K-SVD dictionary learning are utilized to obtain the prominent features of source images. By using the NSCT, the detailed information of source images can be revealed in different scales. Then, using the sparse K-SVD dictionary learning to low-frequency coefficients which are not sparse, salient features of infrared and visible images can be more effectively extracted than other sparse representation methods. Besides, the fourth-order correlation coefficients match strategy is performed to select the suitable high-frequency coefficients to preserve the detailed characteristics of infrared and visible images. The experimental results show that the proposed method outperforms other classical methods.