Journals Proceedings

Abstract

Medical image fusion is defined as the retrieval of complementary information from multiple medical images, for diagnostic purposes and clinical analysis. In our research, 2D discrete wavelet transform (2D-DWT) as an image deomposition method, coupled with principal component analysis (PCA) are applied as a new strategy for fusion of medical CT and MRI images. 2D-DWT is an established method of data decomposition and fusion, whereby it is compact and may provide directional information. The novelty of our work is where the decomposed wavelet images are then processed using PCA-based fusion rule. This approach manages to improve upon corners, edges and contrasts of sources image, thereby providing an altogether higher quality fused image. Simulation results demonstrate high visual quality of the fused image supported by higher values of fusion metrics; these justify the effectiveness of the proposed scheme in comparison to previous methods for image fusion.