Using Deep Learning for Defect Classification on a Small Weld X-ray Image Dataset
- PDF / 2,399,704 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 107 Downloads / 208 Views
Using Deep Learning for Defect Classification on a Small Weld X-ray Image Dataset Chiraz Ajmi1 · Juan Zapata1
· José Javier Martínez-Álvarez1 · Ginés Doménech1 · Ramón Ruiz1
Received: 17 January 2020 / Accepted: 30 August 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract This document provides a comparative evaluation of the performance of a deep learning network for different combinations of parameters and hyper-parameters. Although there are numerous studies that report on performance in deep learning networks for ordinary data sets, their performance on small data sets is much less evaluated. The objective of this work is to demonstrate that such a challenging small data set, such as a welding X-ray image data set, can be trained and evaluated obtaining high precision and that it is possible thanks to data augmentation. In fact, this article shows that data augmentation, also a typical technique in any learning process on a large data set, plus that two image channels, such as channels B (blue) and G (green), both are replaced by the Canny edge map and a binary image provided by an adaptive Gaussian threshold, respectively, gives to the network a 3% increase in accuracy, approximately. In summary, the objective of this work is to present the methodology used and the results obtained to estimate the classification accuracy of three main classes of welding defects obtained on a small set of welding X-ray image data. Keywords Industrial X-ray images · Welding defects · Heterogeneities classification · Deep learning · Machine learning
1 Introduction Welds are customarily used to attach two or more metal parts in a wide range of industrial activities. Because these junctions may suffer loads and fatigue during product lifetime, there is a possibility that they may be deficient due to typical welding defects such as lack of fusion or porosity inside the weld could cause a assemble to break or a structure to rupture. Therefore, it is often necessary to test specific pieces or materials to determine whether the structure is suitable for its designed use. Ideally such testing should be done without damaging the material, piece or structure. Nondestructive testing (NDT) is a wide group of analysis techniques used in science and technology industry to evaluate that the requirements of a material, component or system are satisfied without causing damage over the piece. In fact, NDT have gone from being a simple laboratory curiosity to an essential tool in industry in the last decades. Welds may be tested using NDT techniques such as industrial radiography, indus-
B 1
Juan Zapata [email protected] Dpto. de Electrónica y Tecnología de Computadoras, Universidad Politécnica Cartagena, Cartagena, Spain
trial computed tomography (CT) scanning using X-rays or gamma rays, ultrasonic testing, liquid penetrate testing, or magnetic particle inspection. Although it is one of the oldest techniques of non-destructive inspection, radiography is still accepted as essential for the control of welded
Data Loading...
