Feature extraction using DPSO for medical image fusion based on NSCT

Medical image fusion is a scheme of getting better toned image by coalescing dissimilar modes of images. In this work, a unique method of image fusion is suggested by combining Nonsubsampled Contourlet Transform (NSCT) decomposition through optimization algorithms. Relative analysis of feature extraction is achieved utilizing Darwinian Particle Swarm Optimization (DPSO) which are applied individually to input images and for output fused image to decrease the memory demand. Qualitative analysis is assessed on Entropy, Standard Deviation, PSNR, Average Difference and Cross Correlation. Experimental outcomes revealed that the non-objective standard of clinical images attained by employing DPSO method is superior to PSO.