A new image denoising method using interval-valued intuitionistic fuzzy sets for the removal of impulse noise

Ananthi, V.P. and Balasubramaniam, P. (2016) A new image denoising method using interval-valued intuitionistic fuzzy sets for the removal of impulse noise. Signal Processing, 121. pp. 81-93. ISSN 0165-1684, DOI https://doi.org/10.1016/j.sigpro.2015.10.030.

Full text not available from this repository.
Official URL: https://doi.org/10.1016/j.sigpro.2015.10.030


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.

Item Type: Article
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
Uncontrolled Keywords: Membership function; Impulse noise; Entropy; Fuzzy set; Hesitation degree
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Engineering
Depositing User: Ms. Juhaida Abd Rahim
Date Deposited: 17 Nov 2017 05:40
Last Modified: 17 Nov 2017 05:40
URI: http://eprints.um.edu.my/id/eprint/18304

Actions (login required)

View Item View Item