"Influence areas" as a tool for testing of image restoration methods

  • Authors:
  • Igor V. Gribkov;Peter P. Koltsov;Nikolay V. Kotovich;Alexander A. Kravchenko;Alexander S. Koutsaev;Andrey S. Osipov;Alexey V. Zakharov

  • Affiliations:
  • Russian Academy of Sciences, In Scientific Research Institute for System Studies, Department of Pattern Recognition and Videographic Information Processing;Russian Academy of Sciences, In Scientific Research Institute for System Studies, Department of Pattern Recognition and Videographic Information Processing;Russian Academy of Sciences, In Scientific Research Institute for System Studies, Department of Pattern Recognition and Videographic Information Processing;Russian Academy of Sciences, In Scientific Research Institute for System Studies, Department of Pattern Recognition and Videographic Information Processing;Russian Academy of Sciences, In Scientific Research Institute for System Studies, Department of Pattern Recognition and Videographic Information Processing;Russian Academy of Sciences, In Scientific Research Institute for System Studies, Department of Pattern Recognition and Videographic Information Processing;Russian Academy of Sciences, In Scientific Research Institute for System Studies, Department of Pattern Recognition and Videographic Information Processing

  • Venue:
  • AICT'11 Proceedings of the 2nd international conference on Applied informatics and computing theory
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In the paper we study energy methods for image restoration, when a restored image is obtained after minimization of an integral functional. Although such a functional is global, only few pixels in a small neighborhood of any pixel on the initial image can influence on the corresponding pixel on the restored image. We call this neighborhood an "influence area" and propose a technique for calculation of such areas and their visualization on computer screen. We apply the whole testing technique based upon this approach to a couple of known image restoration methods. The parameters of these methods, ensuring their better performance, are found.