The effects of fusion structures on image fusion performances

Image fusion is becoming one of the hottest techniques in image processing. Many image fusion methods have been developed in a number of applications. This paper mainly discusses the structures of image fusion process, which is classified as hierarchical fusion structure, overall fusion structure, and arbitrary fusion structure. And the effects of such image fusion structures on the performances of image fusion are analyzed. In the experiment, a typical hyperspectral image data set is fused using the same wavelet transform based image fusion technique, but applying different fusion structures. The differences among their fused images are analyzed. The experimental results testify the theoretical analysis that the performances of image fusion techniques are related not only to the fusion algorithm, but also to the fusion structures, and different image fusion structures that produces different fusion performance even using the same image fusion method.