Performance evaluation of an automatic inspection system of weld defects in radiographic images based on neuro-classifiers

In this paper, we describe an automatic system to detect, recognise, and classify welding defects in radiographic images and evaluate the performance for two neuro-classifiers based on an artificial neural network (ANN) and an adaptive-network-based fuzzy inference system (ANFIS). In a first stage, image processing techniques, including noise reduction, contrast enhancement, thresholding, and labelling, were implemented to help in the recognition of weld regions and the detection of defect candidates. In a second stage, a set of 12 geometrical features which characterize the defect shape and orientation was proposed and extracted between defect candidates. In a third stage, we propose a competition between an artificial neural network (ANN) and an adaptive-network-based fuzzy inference system (ANFIS) for weld defect classification. The automatic system of recognition and classification proposed consists in detecting the four main types of weld defects met in practice plus the non-defect type. The results were compared with the aim to know the method that allows the best classification. The correlation coefficients, matrix of confiance, and the acuracy for the ANN and the ANFIS automatic inspection system were determined. The accuracy or the proportion of the total number of predictions that were correct was a value of 78.9% for the ANN and 82.6% for the ANFIS.