Abstract: Noise reduction in material density images is a necessary pre-processing step for the correct interpretation of dual-energy computed tomography (DECT) images and MRI scanned images . In this paper a new method based on a local adaptive processing to reduce noise in DECT and MRI images will be described. Here the comparative study is made based on the filters used for denoising the image. The methods used to describe the systems are Adaptive Neighbourhood Wiener(ANW) filter, prewitt and anisotropic are used among this which one is best method can be determined based on the Peak Signal Noise Ratio (PSNR) value and Mean Squarer Error (MSE) value of the input image for each filters. In this we added Gaussian noise to DECT scanned images and MRI scanned images for the noisy image we apply this filters to denoise the images. Finally by taking the PSNR average value of the each filter we can conclude which filters is best.

Keywords: Noise reduction, Preprocessing, DECT, MRI, prewitt, anisotropic, PSNR, ANW filter.