Comparative Study of Two Proposed Methods for Image Enhancement of Abnormal MRI Brain Images in Spatial and Frequency Domain

Image enhancement is the part of image pre-processing that is used to improve image quality and adjust image information content for further processing and includes techniques such as removing noise or reducing noise levels to a certain extent, adjusting contrast, sharpening the image, etc. In this research, 5 original grayscale images which are abnormal magnetic resonance imaging (MRI) brain images have been processed by using enhancement techniques in two different domains, spatial and frequency. This research described method of enhancement abnormal MRI brain images by using combining spatial domain techniques in MATLAB and method of enhancement abnormal MRI brain images by using combined 2-D stationary wavelet transform for denoising in Wavelet Analyzer technique within MATLAB and filtering in frequency domain technique in MATLAB. Image data used in this research were obtained from Radiopaedia, an educational radiology resource. The best technique to applied on abnormal MRI brain images has been detected by comparing the results of peak signal to noise ratio (PSNR), signal to noise ratio (SNR) and mean square error (MSE). Based on results obtained for PSNR, SNR and MSE measurements, it has been concluded and confirmed that better results for image enhancement of abnormal MRI brain images gives us combining spatial enhancement techniques.