Multi-Modal Medical Image Fusion Using RGB-Principal Component Analysis

Medical Image Fusion is the process of combining multiple images taken by one or more medical imaging modalities. The purpose of medical image fusion is to incorporate structural and functional information of same body organ into a single image. Extensive research efforts have been made in medical image fusion domain to design fusion algorithms for better performance at a low computational cost. Most of the existing algorithms are designed to fuse multi-modal grayscale images. In this paper, we proposed an image fusion algorithm to combine multi-modal color medical images. The proposed algorithm is based on Principal Component Analysis (PCA) and works in two phases. In the first phase, it generates intermediate images after applying PCA on Red, Green and Blue channels of sources images. In the second phase, the final fused image is obtained after applying PCA on pairs of individual color channels of intermediate images. We compared fusion results of the proposed algorithm with the fusion results of existing image fusion algorithms in the same domain using objective image quality assessment method. Comparative analysis depicted that the proposed algorithm performed better than the most of the existing image fusion algorithms.