A Hybrid Approach for Medical Image Fusion using SVD and PCA

The process of combining multiple images into a single one is called image fusion. This facilitates extracting the information from one image instead of collating the information from multiple images. Image fusion reduces the randomness and redundancy and gives better clarity for human perception especially in the field of medical sciences. In order to improve the information contained in an image, we provide a novel technique in this study termed SVDPCA. We choose two distinct types of images for this experiment, namely Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET). Initially singular value decomposition (SVD) is used on the two input images. The final image is created by processing the SVD result with the Principal Component Analysis (PCA) technique. To perform the experiment, we used 30 image sets. The results are compared with other techniques like PCA, SVD and maximum method. The experiments show that the proposed method SVDPCA which combines both methods gives the best result.