Brain CT and MR Image Fusion Framework Based on Stationary Wavelet Transform

In this paper, a technique is proposed for merging of complementary information of two images to obtain a fused image for more suitable human visual perception and for clinical diagnosis. The stationary wavelet transform (SWT) is one of the methods which has proved to be valuable and efficient in image fusion. In this paper, Entropy of square of low frequency subband coefficients is applied. Further, weighted sum modified Laplacian for high frequency subband coefficients is implemented. These two fusion rules are used as activity measure to fuse Computed Tomography (CT) and Magnetic Resonance (MR) images. The proposed method is compared with existing fusion method of averaging low frequency subband and maximum coefficient selection using SWT. The visual and quantitative analysis is done using cross correlation and root mean square error. Both the analysis showed that the proposed method is superior in fusing CT images with T2-weighted MR images.