Image De-noising Using ROF Technique

Sugandha Sinha, Anil Suryavanshi

Abstract: De-noising is the process of removing noise from an image. Additive white Gaussian noise (AWGN) case is the most studied case for de-noising in image processing. A technique implementing with the help of total variation de-noising as the proposed method is described. In this paper, we define the task of image de-noising as follows. We are providing a gray-scale type image for de-noising purpose that has been already degraded with different type of noises that has been further discussed in this paper. The variance of the noise is assumed to be known. The main goal of this technique is to retrieve our original one. Total variation regularization or we can say total variation de-noising is a technique that was originally developed for AWGN image de-noising by Rudin, Osher, and Fatemi [1]. The Total variation regularization/de-noising technique has since been applied to the different kind of other imaging problems. We focus here on the ROF de-noising technique.

Keywords: Total Variation De-Noising, Additive White Gaussian Noise, Speckle Noise, Impulse Random Noise, Salt & Pepper Noise, Fuzzy Filters, Peak to Signal Ratio, Mean Square Error