Structure-aware image fusion

Most existing multi-modal image fusion methods require multi-scale transforms. However, this requirement does not necessarily lead to the fusion result containing the original intensity of source images, and multi-scale transforms need a high computational complexity. In this paper, we tackle the problem of multi-modal image fusion in the spatial domain with a low computational complexity. A salient structure extraction method and a structure-preserving filter are developed to fuse medical images. The developed structure-preserving filter has a property that it recovers small-scale details of the guidance image in the neighborhood of large-scale structures of the input image. Based on the property of the structure-preserving filter, the fusion result is constructed by combining the output of the structure-preserving filter and the source images. Experiments are conducted to demonstrate the effectiveness of the proposed method in comparison with the state-of-the-art approaches in terms of three performance metrics.