A new image denoising method using interval-valued intuitionistic fuzzy sets for the removal of impulse noise
Document Type
Article
Publication Date
1-1-2016
Abstract
Suppressing noise in digital images is more significant in the field of image processing. In this paper, a novel impulse noise detection method is introduced based on fuzzy sets. Generally fuzzy sets are associated with type-1 vagueness, but interval-valued intuitionistic fuzzy sets (IVIFSs) are tied up with type-2 linguistic uncertainty in which the width of the interval represents vagueness. The proposed method investigates image denoising by modeling this vagueness as entropy. An IVIFS for an image is generated by minimizing entropy. Then type-reduced IVIFS is obtained by taking probabilistic sum of the membership interval. Finally, noisy pixels are detected using directional kernels and are filtered using fuzzy filter. Performances are evaluated using mean square error (MSE), peak signal-to-noise ratio (PSNR), mean absolute error (MAE) and structural similarity (SSIM) index. A comparative analysis on the quality of denoised images shows that the proposed technique performs better than several existing median filters.
Keywords
Membership function, Impulse noise, Entropy, Fuzzy set, Hesitation degree
Divisions
fac_eng
Funders
UGC-BSR-Research fellowship in Mathematical Sciences – 2013–2014,Engineering Faculty of the University of Malaya: Grant no. UM.C/625/1/HIR/MOHE/ENG/42
Publication Title
Signal Processing
Volume
121
Publisher
Elsevier