A Remote Sensing Image Fusion Algorithm Based on NMF and Curvelet

摘要

A new method of remote sensing image fusion is proposed based on the both features of NMF(nonnegative matrix factorization) and Curvelet. Firstly, the registered images of multi-spectral and panchromatic are decomposed by curvelet transform to get coefficients of all layers including coarse,fine and detail scale layers. Secondly,NMF is applied to low frequency coefficients(coarse scale layer), to extract the fusion result of low frequency subband coefficients;and use the combination rule of absolute value maximum for coefficient selection from high frequency coefficients (detail and fine scale layer) .Finally, the fused image is obtained by performing the inverse curvelet transform. The experimental results show that the proposed algorithm can effectively obtain more spectral component and higher spatial resolution than those traditional fusing algorithms based on wavelet and curvelet.