Brain Tumor Image De-noising Using Edge Adaptive Total Variation Denoising Algorithm

Abstract

Brain tumor is an abnormality in brain cells caused due to mutations in brain cells. Detection of these tumors is done by using Magnetic Resonance Imaging (MRI) scanning. Researchers have been working on the automated brain tumor detection and classification techniques to assist doctors in diagnosis process. The MRI scans obtained are sometimes affected by noise. To eliminate this noise, image denoising techniques are used. But these techniques remove the noise at the cost of blurring the edges by lowering the resolution and the quality of the image. Retaining the edges present in the brain tumor image is very important for further processing. This paper presents a brain tumor denoising technique by using Edge adaptive total variation. The proposed algorithm analyses the edges present at every pixel while denoising, by using the gradient angle at the pixel. This enhances the performance of the algorithm by retaining the edges of the image while denoising. The algorithm has been compared with existing techniques and has proven to be very effective in removing noise from the image.