Enhanced medical image fusion of PET and CT using NonsubsampledContourlet transform

Fusion finds application in Bio medical imaging. In this paper, the nonsubsampled contourlet transform (NSCT) has been proposed based on a nonsubsampled pyramid structure and nonsubsampled directional filter banks. First, the source images are decomposed into sub images via NSCT and the salient measure is computed. Then the low frequency coefficients are fused based on average rule and the high frequency coefficients by region energy rule. Finally, the fused output image is reconstructed by applying inverse NSCT. For experimental purposes eight sets of PET and CT images are used and the results are compared with fuzzy based triangular membership function using min-sum-mom method. Objective performance is performed on entropy, RMSE, PSNR and correlation coefficient. Experimental results of NSCT medical images demonstrate the superiority of NSCT method compared with fuzzy logic method in terms of visual inspection and objective performance measures.