|
Abstract
|
Article Information:
Application of Multidimensional Chain classifiers to Eddy Current Images for Defect Characterization
S. Shuaib Ahmed, B.P.C. Rao and T. Jayakumar
Corresponding Author: B.P.C. Rao
Submitted: April 07, 2012
Accepted: April 25, 2012
Published: December 15, 2012 |
Abstract:
|
Multidimensional learning problem deals with learning a function that maps a vector of input
features to a vector of class labels. Dependency between the classes is not taken into account while
constructing independent classifiers for each component class of vector. To counteract this limitation,
Chain Classifiers (CC) approach for multidimensional learning is proposed in this study. In this approach,
the information of class dependency is passed along a chain. Radial Basis Functions (RBF) and Support
Vector Machines (SVM) are used as core for CC. Studies on multidimensional dataset of images obtained
from simulated eddy current non-destructive evaluation of a stainless steel plate with sub-surface defects
clearly indicate that the performance of the chain classifier is superior to the independent classifiers.
Key words: Chain classifier, eddy current testing, multidimensional learning, nondestructive evaluation, radial basis function , support vector machines,
|
|
Abstract
|
PDF
|
HTML |
|
Cite this Reference:
S. Shuaib Ahmed, B.P.C. Rao and T. Jayakumar, . Application of Multidimensional Chain classifiers to Eddy Current Images for Defect Characterization. Research Journal of Applied Sciences, Engineering and Technology, (24): 5544-5547.
|
|
|
|
 |
ISSN (Online): 2040-7467
ISSN (Print): 2040-7459 |
 |
| Information |
|
|
|
|
| Sales & Services |
|
|
|
|
