Edge-preserving filters in a boundary options context

  • Simona Moldovanu Department of Computer Science and Engineering, Electrical and Electronics Engineering, Faculty of Control Systems, Computers, Dunarea de Jos University of Galati, Romania
  • Luminița Moraru Department of Chemistry, Physics and Environment, Faculty of Sciences and Environment, Dunarea de Jos University of Galati, Romania
  • Diana Ștefănescu Department of Computer Science and Engineering, Electrical and Electronics Engineering, Faculty of Control Systems, Computers, Dunarea de Jos University of Galati, Galati, Romania
  • Dorin Bibicu Faculty of Economics and Business Administration, Business Adminstration Department, Dunarea de Jos University of Galati, Romania
Keywords: filter, boundary options, correlation index

Abstract

A processed image becomes a degraded image if it loses its main edges. In the current paper we follow up on our previous work by further elaborating on the accordance between the traditional edge filter outcomes and boundary options operations. The correlation index allows the estimating of the similarity between a raw image and a processed image. The images are processed using the Sobel, Canny, Prewitt and Roberts filters. The same images are also processed with symmetric, circular and replicate boundary options operators. A comparative analysis is performed in order to corroborate the results. The possible optimal combination between the filter action and boundary options is identified.

Downloads

Download data is not yet available.
Published
2017-06-11
How to Cite
Moldovanu, S., Moraru, L., Ștefănescu, D. and Bibicu, D. (2017) “Edge-preserving filters in a boundary options context”, Analele Universității ”Dunărea de Jos” din Galați. Fascicula II, Matematică, fizică, mecanică teoretică / Annals of the ”Dunarea de Jos” University of Galati. Fascicle II, Mathematics, Physics, Theoretical Mechanics, 40(1), pp. 5-11. Available at: https://gup.ugal.ro/ugaljournals/index.php/math/article/view/1236 (Accessed: 28November2024).
Issue
Vol 40 No 1 (2017)
Section
Articles