In multiresolution analysis, the total structures of images are characterized by an approximate component, and the specific features such as lines, edges, and contours are contained in detail components. This paper proposes a fusion scheme to combine the structure information representing the background of source images and the directive contrast information representing the features of lines, edges, and contours along different directions in source images into a single image. Here, the structure information can be attained by structure similarity methods. And the directive contrast is defined as the ratio of the local maxima of the detail components in a Gaussian window and the mean of the corresponding approximate components. The visual and statistical results show that the proposed method has improved the fusion performance compared to the existing fusion schemes, especially in term of the visual effects.