26. Image Fusion Method Based on Sparse and Redundant Representation

In order to obtain the fusion image which can provide more information, a new method of image fusion based on sparse and redundant representation theory is put forward. In the method, first of all, the original image is represented by a redundant dictionary as a sparse coefficient. Then, the sparse coefficients are fused according to the absolute-max fusion rule. Lastly, the fused image is reconstructed based on the merged coefficients and the redundant dictionary. The method proposed in this paper is compared with some traditional methods on some space targets images. They are laplace pyramid fusion method, principal component analysis fusion method, discrete wavelet transform fusion method, curvelet transform fusion method, and non-subsampling contourlet transform fusion method. The experimental results show that the proposed method has better performance both subjectively and objectively.