Journal of Contemporary Issues in Business and Government

Abstract

In this paper, a new Fuzzy based Fast Non Local Mean algorithm is proposed
to denoise Rician noise from MRI images. Initially, Fuzzy function used to find the similar
and non-similar pixel. After this, Non Local Mean(NLM) algorithm with integral
image representation is used to find the weights of similar pixel at a faster rate than the
normal NLM algorithm in the image. Consequently these similar pixels are used to generate
the noise free pixels. At the end the conventional bias subtraction method is used as
post processing step. The proposed scheme is tested with real data set and compared with
existing Fast NLM techniques and basic NLM using Root mean square error(RMSE),
peak signal noise ratio (PSNR), Structure similar index (SSIM), quality index and computational
time parameters methods. The proposed method gave better result than existing
Fast NLM technique with high and density Rician noise in the image and it is Fast
than NLM.