Denoising of images using fuzzy rulebase and clustering approach
Saikat Maity, Jaya Sil · 2014
Detection of pixels corrupted by noise and assessing the degree to which the pixels are corrupted intrinsically fuzzy processes, involve uncertainty and imprecision. The paper aims at reconstruction of the image with ensured quality after removing noise from the original image. Here region marking process has been introduced to obtain number of clusters automatically which partition the whole image into several regions. Noise pixels in each region are detected by applying fuzzy rules constructed using gradient information of the image. The proposed approach reproduces the image with promising quality compare to the existing image denoising methods face. It also gives a comparative study between the existing filters with the proposed one not only for grey scale images but also the colour images.