Evaluation and detection of defects of industrial surfaces and welds, using radiographic images and Euler's number

  • Authors:
  • Alireza Zendebudi;Seyed Alireza Hashemi

  • Affiliations:
  • Telecommunication company of Bushehr and Azad Islamic University of Bushehr;Telecommunication company of Bushehr and Azad Islamic University of Bushehr

  • Venue:
  • ISPRA'10 Proceedings of the 9th WSEAS international conference on Signal processing, robotics and automation
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Detection of industrial defects such as cracks and holes on outer surfaces of industrial equipments in the oil and gas industry is much important. Traditional methods of visual inspection of reservoirs, and, in some cases, some radiographic images prepared using x-ray projection machines, which have a high depreciation rate in the industries at the present, show no good precision and accuracy. In this article, we use the relative efficiency of the binary algorithm and the binary morphology algorithm on the images in question, to give a new evaluation of the likelihood of presence of defects on outer surfaces of welded equipments in the gas and oil field. In view of the fact that if the quality of the grey radiographic images prepared is not ok, their gray image algorithm, their histogram and computation of their first and second moment computations have no suitable and acceptable applications, therefore, replacing them with a more accurate method seems necessary. In this study, different samples of radiographic images prepared from among as image are selected and Euler's numerical computations with be carried out for three kinds of radiographic image with different defects. There, by comparing the two algorithms mentioned to and evaluation of their efficiency in defecting void and crack defects, the ability of the method based on Euler's number is determined. The findings and evaluations show that the binary morphology method has a higher ability and accuracy in detecting defects.