Automatic fuzzy based identification of flawed rivet in Magneto Optic Images

Issue title: Selected Papers from the 13th International Symposium on Applied Electromagnetics and Mechanics (ISEM 2007)

Article type: Research Article

Authors: Cacciola, Matteo^{a; *} | Deng, Yiming^b | Morabito, Francesco C.^a | Udpa, Lalita^b | Udpa, Satish^b | Versaci, Mario^a

Affiliations: [a] Universitá degli Studi di ReggIo Calabria "Mediterranea", DIMET, Via Graziella Feo di Vito, 89100 Reggio Calabria, Italy | [b] Michigan State University, College of Engineering, NDEL, East Lansing, MI 48824, USA

Correspondence: [*] Corresponding author. E-mail: matteo.cacciola@unirc.it

Abstract: Magneto Optic Imaging is a relatively recent Non Destructive Evaluation technique, successfully used to evaluate the integrity of metallic parts in aging aircrafts. In this paper, the authors deal with the implementation of a suitable procedure able to recognize flawed rivets in Magneto Optic measurements. They have been carried out at the NDE Lab, Michigan State University, USA. Exploiting Fuzzy Subtractive Clustering in order to implement the classifier, our proposed approach retrieves a fast and robust applicative, useful for real-time applications. Preliminary results are really encouraging.

Keywords: Magneto optic images, fuzzy clustering, flawed rivet identification

DOI: 10.3233/JAE-2008-988

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 28, no. 1-2, pp. 297-303, 2008