Image Enhancement Techniques Applied to Magnetic Resonance Images: Multiple sclerosis

Abstract

As the most crucial step in image preprocessing techniques, Contrast enhancement (CE) is utilized to improve low contrast images quality with the aim to ameliorate the image’s visual appearance for later automated processing (detection, segmentation and recognition). Due to this importance, several techniques were proposed to treat this topic. The present manuscript proposes comparative research across some contrast enhancement methods mainly: Global Histogram Equalization (GHE), Adaptive Gamma Correction (AGC) and knee function and gamma correction based on Singular Value Decomposition with Discrete Wavelet Transform (SVD-DWT) method (KGC-DWT-SVD). These methods had been applied on brain and spinal cord MR (Magnetic Resonance) images of patients affected with multiple sclerosis. The obtained results had been evaluated according to the following measurement metrics: the entropy (H), peak signal to noise ratio (PSNR), mean squared error (MSE), structure similarity index measurement (SSIM) and feature similarity index measurement (FSIM).