Abstract: Effective technique for denoising is necessary for medical images particularly in Computed Tomography, which is a significant and most general modality in medical imaging. In this work denoising of Computed Tomography (CT) images is performed which generally gets degraded by the presence of Additive White Gaussian Noise (AWGN). This work presents three techniques to denoise CT images namely Discrete Wavelet Transform (DWT) Thresholding, Non-local (NL) means filter and wavelet thresholding and joint NL means filtering. In this report, we explore wavelet denoising of images using several thresholding techniques such as SURE Shrink, Visu Shrink, Bayes Shrink and Neigh Shrink. NL-means filter assumes that the image contains an extensive amount of redundancy and therefore it uses the concept of self-similarity in an image to denoise it. The third technique is a combination of the above two methods. Data evaluations are accomplished by using two criterions; namely, peak signal to noise ratio (PSNR) and mean square error (MSE). It is observed that NL means performs better Wavelet denoising.

Keywords: CT image denoising, AWGN, Wavelet Thresholding, NL means, PSNR, Computation time.