Wavelet Coefficients Thresholding Techniques for Denoising MRI Images

Background: Image denoising is one of the primary challenges in the field of image processing. The objective is to derive the original image by suppressing noise from a noisy the image. The need of image denoising techniques is still in high demand. De-noising of medical images is degraded by various noises. Multiresolution techniques are very efficient for medical image denoising. Methodology: In this work, it is proposed to examine the efficiency of different wavelet shrinkage thresholding techniques and to determine the best one. Findings: The metric used for analysis are PSNR, VSNR and WSNR. The experimental results show that third level decomposition of Symlet in association with Neigh Shrink threshold outperforms all other approaches. Applications/Improvements: In this paper, denoising is applied to MRI images. Gaussian Noise, Salt and Pepper Noise, and Speckle Noise can be removed using the methods mentioned. The methods can also be extended to denoising other medical images like CT scan, X-RAY, and Ultra Sound etc.