An Improved Image Fusion Method of Infrared Image and SAR Image via Shearlet and Sparse Representation

摘要

In this paper, a novel image fusion method of infrared image and SAR image is proposed combining Shearlet transform and sparse representation to avoid the disadvantages of Wavelet transform. The registered images are decomposed by the shearlet transform to obtain the low frequency subband and a series of high frequency subbands. The low frequency subband with lower sparseness is disposed with sparse representation, construct over complete dictionary, solve sparse coefficient over the trained dictionary, and choose the low frequency coefficients with the larger energy fusion rule. And the rule of gradient absolute value maximization is applied to the high frequency subbands. Then the fusion image is obtained by the inverse Shearlet transform. Experimental results show that the proposed method can retain good visual quality and objective evaluation index, and performs some related fusion approaches.