Filtering Random Valued Impulse Noise from Grayscale Images through Support Vector Machine and Markov Chain

Gellert Arpad (), Brad Remus (), Morariu Daniel () and Neghina Mihai ()
Additional contact information
Gellert Arpad: Computer Science and Electrical and Electronics Engineering Department, Faculty of Engineering, “Lucian Blaga” University of Sibiu, Romania
Brad Remus: Computer Science and Electrical and Electronics Engineering Department, Faculty of Engineering, “Lucian Blaga” University of Sibiu, Romania
Morariu Daniel: Computer Science and Electrical and Electronics Engineering Department, Faculty of Engineering, “Lucian Blaga” University of Sibiu, Romania
Neghina Mihai: Computer Science and Electrical and Electronics Engineering Department, Faculty of Engineering, “Lucian Blaga” University of Sibiu, Romania

International Journal of Advanced Statistics and IT&C for Economics and Life Sciences, 2021, vol. 11, issue 1, 70-84

Abstract: This paper presents a context-based filter to denoise grayscale images affected by random valued impulse noise. A support vector machine classifier is used for noise detection and two Markov filter variants are evaluated for their denoising capacity. The classifier needs to be trained on a set of training images. The experiments performed on another set of test images have shown that the support vector machine with the radial basis function kernel combined with the Markov+ filter is the best configuration, providing the highest noise detection accuracy. Our filter was compared with existing denoising methods, it being better on some images and comparable with them on others.

Keywords: random valued impulse noise; support vector machine; Markov filter; denoising (search for similar items in EconPapers)
Date: 2021
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://doi.org/10.2478/ijasitels-2021-0004 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:vrs:ijsiel:v:11:y:2021:i:1:p:70-84:n:7

DOI: 10.2478/ijasitels-2021-0004

Access Statistics for this article

International Journal of Advanced Statistics and IT&C for Economics and Life Sciences is currently edited by Daniel Volovici

More articles in International Journal of Advanced Statistics and IT&C for Economics and Life Sciences from Sciendo
Bibliographic data for series maintained by Peter Golla ().