N. Nacereddine, R. Draï, and A. Benchaâla (Algeria)
Weld defects, Extraction, Classification, Neural Networks.
In the first part of the paper, we show the effectiveness of using neural network paradigms to segment in edges the weld defect images obtained by industrial radiography. The developed classifier consists of a multilayer feed forward network window in which the center pixel was classified using gray scale information within the window. The aim of the work in the second part is to construct a set of weld defect descriptors in X-ray images and then classify them by the neural classifier. These descriptors are based on the geometric invariant moments, which are insensitive regarding usual geometric transformations. Once the geometric invariant features computed, a neural network trained by back-propagation classifies the defect-images in planer or volumetric defect classes.
Important Links:
Go Back
