A novel method for medical image fusion using modified non-sub sampled contour let transform

Abstract

In this paper, a novel multimodal medical image fusion is proposed based on Non-sub sampled contourlet transform (NSCT), which almost transfers the information from source images to fused image. Initially, the source images are subjected to NSCT followed by the fusion of low frequency sub images and high frequency sub images. The low frequency sub images are fused based on the phase congruency rule where as the high frequency sub images are fused based on the dependencies of Log-Gabor energies. The proposed approach is compared with conventional approaches such as DWT, Contourlet Transform (CT), and some approaches based on NSCT over various test images. Performance evaluation is carried out visually and numerically. Both the visual and numerical results illustrate the effectiveness of proposed approach and also show the enhancement of proposed approach over conventional approaches.