Multi-modal image fusion using contourlet and wavelet transforms: a multi-resolution approach

Abstract

In recent years, vast improvement and progress has been observed in the field of medical research, especially in digital medical imaging technology. Medical image fusion has been widely used in clinical diagnosis to get valuable information from different modalities of medical images to enhance its quality by fusing images like computed tomography (CT), and magnetic resonance imaging (MRI). MRI gives clear information on delicate tissue while CT gives details about denser tissues. A multi-resolution approach is proposed in this work for fusing medical images using non-sub-sampled contourlet transform (NSCT) and discrete wavelet transform (DWT). In this approach, initially the input images are decomposed using DWT at 4 levels and NSCT at 2 levels which helps to protect the vital data from the source images. This work shows significant enhancement in pixel clarity and preserves the information at the corners and edges of the fused image without any data loss. The proposed methodology with an improved entropy and mutual information helps the doctors in better clinical diagnosis of brain diseases.