Multilevel medical image fusion using multi-level local extrema and non sub-sampled contourlet transformation

In recent days, Multimodal image fusion is one of the important aspects especially in clinical diagnosis applications. Multimodal image fusion fuses two images obtained from the different imaging devices. This paper proposed a two-stage multimodal image fusion framework using the parallel combination of Multilevel Local Extrema (MLE) and Non Sub-Sampled Contourlet Transform (NSCT). Furthermore to improve the shift variance, directionality, and phase information in the fused image using the NSCT technique. The performance of the proposed work is tested with six different standard data set images. The experimental results shows that the performance of proposed method superior than several existing state-of-art methods in terms of quality metrics.