Effect of image transformations on dynamic stochastic resonance based MR image enhancement

Abstract

Dynamic Stochastic Resonance (DSR) utilizes the noise associated with the image itself to enhance the image quality. This paper analyzes the effects of Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT) and Singular Value Decomposition (SVD) of image, which works as the input to bi-stable nonlinear system exhibiting DSR. The study performed on T1, fluid-attenuated inversion recovery (FLAIR) and diffusion weighted sequences of Magnetic Resonance Imaging (MRI). The images were quantified in terms of contrast enhancement factor and image anisotropy. The results show that DSR based image enhancement is helpful to obtain better tissue differentiation. The DCT based DSR produces better enhancement for diffusion-weighted images whereas DWT and SVD based DSR produces better enhancement of T1 and FLAIR weighted magnetic resonance images. © 2017 IEEE.