A new image denoising model utilizing the conformable fractional calculus for multiplicative noise

Abstract

Reducing noise from images is an essential structure of the image processing study. Noises can arise with images through achievement on diffusion. The existence of noise can delay the right operation of these images for many applications such as satellite and medical images. Reducing denois in images multiplicatively (DIM) has been developed and modified by many researchers during the past few years. DIM can destroy almost all data of the original image, especially the texture of images. Our aim is to present a new technique to solve this problem. The technique is based on a new fractional calculus called the conformable fractional calculus (CFC). This type of calculus has advantages because of its formula involves a controller, which can be applied to complex problems such as DIM. The proposed structures of CFC windows are given by four masks suggested for x and y directions. On four directional angles, a convolution operational product of the input image pixels with a CFC mask window has been completed. The visual observation and peak signal-to-noise ratio with Root Mean Square Error are employed for measurements. The experiments showed that the skillful filtering outcomes are indicated high score than some well known filers such as Gaussian filter, Sobel edge filter, Canny edge filter and gray-level co-occurrence matrix. Compering is illustrated as well with newly researches.