Effective de-noising and characterisation of defects in eddy current-NDE images
by Balakrishnan Sasi; M.S. Ahamed Ali; C. Babu Rao; T. Jayakumar
International Journal of Signal and Imaging Systems Engineering (IJSISE), Vol. 8, No. 4, 2015

Abstract: Non Destructive Evaluation (NDE) plays an important role in assessment of structural integrity of engineering components. Among various NDE techniques, eddy current NDE is widely used for detection of defects in installed components because of its ease of operation, non-contact and versatility. However, in the case of welded components, the presence of microstructural variations affects the material parameters such as conductivity and permeability. This in-turn produces large amplitude disturbing signals (weld noise) that mask small amplitude signals from defects. This demands the development of de-noising methodology, which enhances the characterisation of defects using Artificial Neural Network (ANN). This paper discusses the discrete wavelet transform-based de-noising method for elimination of weld noise and enhancement of defect information. The paper also briefs the characterisation of defects from the enhanced defect region using Multilayer Perceptron (MLP) based back propagation ANN.

Online publication date: Fri, 10-Jul-2015

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Signal and Imaging Systems Engineering (IJSISE):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com