Bayesian Networks-Based Defects Classes Discrimination in Weld Radiographic Images

Type : Article de conférence
Auteur(s) :  Aicha Baya Goumeidane, Abdessalmen Bouzaeini, Nafaa Nacereddine, Salvatore Tabbone
Année :  2015
Domaine : Electronique
Conférence: 16th Computer Analysis of Images and Patterns (CAIP 2015)
Résumé en PDF :  (résumé en pdf)
Fulltext en PDF : 
Mots clés :  Bayesian networks, weld defects, Geometric descriptors, radiography

Résumé : 

Bayesian (also called Belief) Networks (BN) is a powerful knowledge representation and reasoningmechanism. Based on probability theory involving a graphical structure and random variables, BN iswidely used for classification tasks and in this paper, BN is used as a class discrimination tool for a set ofweld defects radiographic images using suitable attributes based on invariant geometric descriptors. Testsare performed on a database of few hundred elements where the results are outstanding and verypromising, since they outperform those given by powerful SVM classifiers.