Impulsive noise suppression from highly distorted images with triangular interpolants

Civicioglu P., Alci M.

AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS, vol.58, no.5, pp.311-318, 2004 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 58 Issue: 5
  • Publication Date: 2004
  • Doi Number: 10.1078/1434-8411-54100250
  • Journal Name: AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.311-318
  • Keywords: impulsive noise, triangular interpolants, scattered data interpolation, salt-and-pepper noise, WEIGHTED MEDIAN FILTERS, CORRUPTED IMAGES, REMOVAL
  • Erciyes University Affiliated: Yes

Abstract

In this paper a new filter, Triangular Interpolant Based Impulsive Noise Suppression Filter (SF), is proposed to restore images corrupted by fixed valued impulsive noise (IN). The proposed filter comprises two main stages: detection of noise and restoration of corrupted pixels. The SF achieves the restoration of the detected noisy pixels by using one of the Triangular Interpolant techniques and leaves the other pixels unaltered. Simulation results reveal that the proposed filter shows better performance than the highly approved IN suppression filters across a wide noise density ranging from 10% to 90%. The proposed filter also perfectly achieves the robustness and detail preservation with reduced computational complexity.