Multimodal Image Fusion Using Curvelet and Genetic Algorithm

Fusion of medical images of different modalities always have the advantages in efficient medical diagnosis. Magnetic resonance image (MRI) and Computed tomography (CT) are twp such modalities which are generally fused. The existing fusion techniques like wavelet transformation have proved to be good in medical image fusion. However, they have failed to retain certain quality with respect to the original. In this paper, one such attempt is made by combining the popular Curvelet transformation (CTr) with Genetic Algorithm (GA). The performance of the proposed method is evaluated in terms of PSNR and MSE while fusing MRI and CT of brain. The results clearly mentioned that the Curvelet and the GA-CTr combination have better fusion characteristics than the WT.