OSA's Digital Library

Journal of the Optical Society of America A

Journal of the Optical Society of America A

| OPTICS, IMAGE SCIENCE, AND VISION

  • Editor: Franco Gori
  • Vol. 31, Iss. 1 — Jan. 1, 2014
  • pp: 196–205

Defect detection of castings in radiography images using a robust statistical feature

Xinyue Zhao, Zaixing He, and Shuyou Zhang  »View Author Affiliations


JOSA A, Vol. 31, Issue 1, pp. 196-205 (2014)
http://dx.doi.org/10.1364/JOSAA.31.000196


View Full Text Article

Enhanced HTML    Acrobat PDF (4028 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

One of the most commonly used optical methods for defect detection is radiographic inspection. Compared with methods that extract defects directly from the radiography image, model-based methods deal with the case of an object with complex structure well. However, detection of small low-contrast defects in nonuniformly illuminated images is still a major challenge for them. In this paper, we present a new method based on the grayscale arranging pairs (GAP) feature to detect casting defects in radiography images automatically. First, a model is built using pixel pairs with a stable intensity relationship based on the GAP feature from previously acquired images. Second, defects can be extracted by comparing the difference of intensity-difference signs between the input image and the model statistically. The robustness of the proposed method to noise and illumination variations has been verified on casting radioscopic images with defects. The experimental results showed that the average computation time of the proposed method in the testing stage is 28 ms per image on a computer with a Pentium Core 2 Duo 3.00 GHz processor. For the comparison, we also evaluated the performance of the proposed method as well as that of the mixture-of-Gaussian-based and crossing line profile methods. The proposed method achieved 2.7% and 2.0% false negative rates in the noise and illumination variation experiments, respectively.

© 2013 Optical Society of America

OCIS Codes
(100.2960) Image processing : Image analysis
(150.1835) Machine vision : Defect understanding

ToC Category:
Machine Vision

History
Original Manuscript: June 28, 2013
Revised Manuscript: October 25, 2013
Manuscript Accepted: November 27, 2013
Published: December 24, 2013

Citation
Xinyue Zhao, Zaixing He, and Shuyou Zhang, "Defect detection of castings in radiography images using a robust statistical feature," J. Opt. Soc. Am. A 31, 196-205 (2014)
http://www.opticsinfobase.org/josaa/abstract.cfm?URI=josaa-31-1-196

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

You do not have subscription access to this journal. Article level metrics are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Log in to access OSA Member Subscription

« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited