Medical image fusion techniques based on combined discrete transform domains

This paper investigates some of medical image fusion techniques and discusses the most important advantages and disadvantages of these techniques to develop hybrid techniques that enhance the fused image quality. Both traditional and hybrid fusion algorithms are evaluated using several quality metrics including average gradient, local contrast, standard deviation, edge intensity, entropy, structure similarity index, universal image quality index, feature similarity index, Peak Signal-to-Noise Ratio (PSNR), mutual information, Q
^ab/f
, and processing time. Experimental results prove that the hybrid technique of Additive Wavelet Transform (AWT) and Dual Tree complex wavelet transform (DT-CWT) with high pass sharpening filter provides the best fused images of highest quality, highest details, shortest processing time, and best visualization. This is favourable, especially for helping in accurate diagnosis and optimal therapy applications.