Share Email Print

Proceedings Paper

Using a microcomputer in fault detection
Author(s): Kun-Li Wen; John H. Wu
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

X-Ray radiography method is a typicial human-visual testing among NDT (nondestructive testing). Due to the labor's film-identified needs and the long practical experience, the skilled labors are hard to be trained. Not only the labor's skiliness and spiritness will affect the quality of the film-identified, hut also the qualtity parameters of flaw (e.g. the size of flaw) cannot be determined by labor within short time. However, computer vision image processing system can give some good characteristics ,such as, high speed, quantitative parameters and non-human's error etc. Developing this system to assist the labor's film-identified will be certainly assuring the film quality, meanwhile, it will be the most powerful method of on-line flaw testing in the future. This paper just focuses the research topic at the identification of X-ray film for the butt welding steel materials1. First, to analize the defect's image model in the X-ray film, then by the image processing technique to build up the propper edge detecting operator and the edge detecting rule, and finally, by the derived edge detector operator to do the mask operation to the X-ray film image, and to detect the flow contour from the segmented defect image for following identification and classification. In this study, we make use of the fuzzy pattern recognition2 and hierarchy classifier to identify the welding flaws.

Paper Details

Date Published: 22 September 1993
PDF: 11 pages
Proc. SPIE 2101, Measurement Technology and Intelligent Instruments, (22 September 1993);
Show Author Affiliations
Kun-Li Wen, National Central Univ. (Taiwan)
John H. Wu, National Central Univ. (Taiwan)

Published in SPIE Proceedings Vol. 2101:
Measurement Technology and Intelligent Instruments
Li Zhu, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?