A study of two image-recognition algorithms for the classification of flaws in a test object according to its digital image

S. E. Vorobeichikov, V. A. Fokin, Victor Anatolievich Udod, A. K. Temnik

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The mathematical model of a digital radiation image is given that corresponds to a test object that contains two types of defects, namely, pores and cracks (or non-penetrating defects). Two algorithms for automatic image recognition of these defects on a digital radiation image of an object are given. The effectiveness of the algorithms was evaluated via mathematical simulation.

Original languageEnglish
Pages (from-to)644-651
Number of pages8
JournalRussian Journal of Nondestructive Testing
Volume51
Issue number10
DOIs
Publication statusPublished - 1 Oct 2015

    Fingerprint

Keywords

  • defects
  • digital radiation image
  • recognition algorithms
  • test object

ASJC Scopus subject areas

  • Materials Science(all)
  • Condensed Matter Physics
  • Mechanics of Materials
  • Mechanical Engineering

Cite this