<p>This work introduces a new method for multi-focal and medical image fusion applications. The process is based on multi-scale decomposition (MSD) criteria in which filtering at multiple levels with focus detection is applied. The MSD tool used in this algorithm is Hilbert vibration decomposition (HVD), which decomposes an image into a set of instantaneous image amplitudes. These image amplitudes are passed through a mean filter to obtain filtered image amplitudes. Then, with the help of a fundamental difference operator, we get the high-frequency information, i.e., rough focus maps. These rough focus maps are further refined using guided filtering to obtain accurate focus maps. After this, weights are estimated on the choose max rule to generate the initial instantaneous fused image amplitudes. For spatial consistency verification, these fused image amplitudes are again filtered with the help of a guided filter. Then, the final weights are used to obtain the absolute instantaneous fused image amplitudes. These are recombined to generate the output fused image. In addition to this, we also apply a simplified image enhancement approach at the preliminary stage for better textures and contrast at the output. The simulation results obtained with this approach are better than those obtained with the existing methods.</p>