Multimodal Medical Image Fusion Using DC Coefficient Scaling and MWGF in DWT Domain

Abstract

 Multimodal pictures, as the term implies, are medical images that were obtained with the 
use of numerous camera modules. A difficult challenge awaits us in extracting these photos' 
characteristics. As a result, both photos have a vastly different amount of information and low 
contrast when compared to each other. These photos have a poor contrast because of a lack of 
doctor's competence and practical hands. There are two major contributions to this study. This 
approach uses two layers of DWT, and then a modified DC coefficient scaling method in compressed 
DCT domain to improve multimodal pictures with low contrast. L component DC coefficient is 
adjusted to improve entropy and brightness in translated CIE-Lab colour space. It is then utilised to 
merge the two improved pictures by using a Modified Weighted Gradient Fusion (MWGF). For the 
final reconstruction, the inverse DWT is applied to the fused picture. Images from a variety of 
environments are used to evaluate the approaches.