Medical image fusion by combining parallel features on multi-scale local extrema scheme

Two efficient image fusion algorithms are proposed for constructing a fused image through combining parallel features on multi-scale local extrema scheme. Firstly, the source image is decomposed into a series of smoothed and detailed images at different scales by local extrema scheme. Secondly, the parallel features of edge and color are extracted to get the saliency maps. The edge saliency weighted map aims to preserve the structural information using Canny edge detection operator; Meanwhile, the color saliency weighted map works for extracting the color and luminance information by context-aware operator. Thirdly, the average and weighted average schemes are used as the fusion rules for grouping the coefficients of weighted maps obtained from smoothed and detailed images. Finally, the fused image is reconstructed by the fused smoothed and the fused detailed images. Experimental results demonstrate that the proposed algorithms show the best performances among the other fusion methods in the domain of MRI-CT and MRI-PET fusion.