Scientific Research and Essays

  • Abbreviation: Sci. Res. Essays
  • Language: English
  • ISSN: 1992-2248
  • DOI: 10.5897/SRE
  • Start Year: 2006
  • Published Articles: 2754

Full Length Research Paper

Edge detection via statistical features and automatic thresholding technique

Salim Ben Chaabane
  • Salim Ben Chaabane
  • SICISI Unit, ESSTT, 5 Av. Taha Hussein, 1008, Tunis, Tunisia.
  • Google Scholar
Mounir Sayadi
  • Mounir Sayadi
  • SICISI Unit, ESSTT, 5 Av. Taha Hussein, 1008, Tunis, Tunisia.
  • Google Scholar
Farhat Fnaiech
  • Farhat Fnaiech
  • SICISI Unit, ESSTT, 5 Av. Taha Hussein, 1008, Tunis, Tunisia.
  • Google Scholar


  •  Accepted: 21 December 2012
  •  Published: 31 January 2013

Abstract

In this paper, the problem of edge detection is addressed using the first order statistics and automatic thresholding technique. The general idea of edge detection using the simple edge detectors such as gradient operators or second derivative operators is extended to the statistic domain. The statistical features are used to describe the relationship between the current pixel and its neighboring, then, the thresholding technique is employed to determine the edge of gray level image. The proposed method improves the accuracy of the edge detection and suppresses the impact of the noise on the results, while the edge has a good consistency. The proposed method is validated by performing a comparative study with respect to other existing techniques. The experimental segmentation results, on standard and textured images, highlight the effectiveness of the proposed method.

 

Key words: Thresholding, statistical features, first order statistics, noise, segmentation, edge detection, defect detection.