A Remote Sensing Image Fusion Algorithm Based on the Second Generation Curvelet Transform and DS Evidence Theory

Abstract

A new method of remote sensing image fusion is proposed based on the second generation Curvelet transform and Dempster-Shafer (DS) evidence theory. In this paper, the remote sensing images are decomposed by the Curvelet transform to get the coefficients and optimize the high coefficients with DS evidence theory. Firstly, the high resolution and multispectral remote sensing images are decomposed by the Curvelet transform to get the Curvelet transform coefficients of all layers (Coarse, Detail and Fine scale layer). Secondly, the Coarse scale layer is used the maximum fusion rule. The Detail scale layer is used by the weighted average fusion rule. The Fine scale layer is optimized by the DS evidence theory. Get the three features of the Fine scale layer coefficients. The three features are the variance, information entropy and energy. Use the features to be some parameters belief function and the plausibility function. Then compose the mass function and get new fusion coefficients. Finally, the fused image is obtained by the inverse Curvelet transform. The experimental results show that the new algorithm can more effectively than wavelet and other traditional fusion algorithms such as HIS, brovey in the remote sensing image fusion.