Medical image noise reduction using Radon transform and Walsh list in Laplacian pyramid domain

Abstract

It is well known that noise on medical image resulting in low image quality has largely limited the diagnostic usefulness. Therefore, noise reduction for medical images is significantly urgent. However, most reported noise reduction methods, which are usually based on the local statistics of a medical image, are not efficient for medical image noise reduction. This paper presents a novel approach for medical image noise reduction using Radon transform (RT) and an adaptive median filter based on Walsh list in the Laplacian pyramid domain. After that, experiments results on artificial and clinical X-ray computed tomography (CT) images with Gaussian noise and salt and pepper noise using proposed method and traditional methods have been summarized. Also, speckle reduction on ultrasound (US) image has been performed using artificial US image. Based on the experimental results, it is concluded that the proposed method showed better capability for noise reduction on medical images compared with other traditional methods.