Edge-guided filtering based CT image denoising using fractional order total variation

CT scans, which are another name for computed tomography, are one of the many diagnostic tools where the use of photon -counting allows for the differentiation of materials and the characterization of tissues. The CT image quality may be compromised in terms of noise and artifacts due to low radiation dose. CT scans make it easier for radiologists to find important diagnostic information. So, keeping a noiseless medical image is important if you want to collect the most important medical information. But a number of studies show that noise is a common problem with CT images, especially in low contrast images where it can hide important medical information. This is worrying because noise can make it hard to figure out what the data means. To overcome these issues, this paper introduces a new weighted function which is utilized by fractional order total variation for CT image denoising and resolve the problems of the blocky effect, also to resolve nonconvex optimization problem for better solution. Two different methods (i) Split Bregman, and (ii) Augmented Lagrangian, were examined by utilizing proposed weighted fractional total variation denoising. The numerical experimental evaluations and comparative study are also performed for validate and evaluate the performance of proposed method.