摘要

A multimodal medical image fusion method based on discrete fractional wavelet (DFRWT) is presented in this paper. With a change in p order in domain (0,1], source medical images are decomposed by DFRWT in different p order. The sparsity character of the mode coefficients in subband images changes. According to the method, to enhance the correlation between subband coefficients, the non-sparsity character of the mode coefficients in low p order should be utilized. The coefficients of the all subbands are fused using the weighted regional variance rule. Finally, inverse DFRWT is applied to obtain a fused image. Subjective and objective analyses of the results and comparisons with other multiresolution domain techniques show the effectiveness of the proposed scheme in fusing multimodal medical images.