Title: Multi-weld defects detection based on Gabor filter, Hough transform

Authors: Chiraz Ajmi; Sabra Elferchichi; Juan Zapata; Abderrahmen Zaafouri; Kaouther Laabidi

Addresses: Departamento Tecnologías de la información y las comunicaciones, Universidad Politécnica de Cartagena, Campus la Muralla, Edif. Antigones, 30202 – Cartagena (Murcia), Spain ' University of Jeddah, Jeddah, 21959, Saudi Arabia ' Departamento Tecnologías de la información y las comunicaciones, Universidad Politécnica de Cartagena, Campus la Muralla, Edif. Antigones, 30202 – Cartagena (Murcia), Spain ' National Superior School of Engineering of Tunis, University of Tunis, Street Taha Hussein, 1008, Tunisia ' CEN Department, University of Jeddah, Jeddah, 21959, Saudi Arabia

Abstract: Weld defect detection is an important application in the field of non-destructive testing (NDT). These defects are mainly due to manufacturing errors or welding processes. In this context, image processing especially segmentation is proposed to detect and localise efficiently different types of defects. It is a challenging task since radiographic images have deficient contrast, poor quality and uneven illumination caused by the inspection techniques. The usual segmentation technique uses a region of interest (ROI) from the original image. A robust and automatic method is presented to detect two major defect types from mono or multi-weld defects images. So, pre-processing tools are applied based on Gaussian filter and contrast stretching then segmentation too is performed based on Gabor filter, binarisation and Canny detector to extract edges and finally detection and location of multi-weld defects with a modified 'Hough transform' technique. The experimental results show that our proposed method gives good performance.

Keywords: weld defect; radiography; NDT; non-destructive testing techniques; Hough transform; Canny detector; adaptative thresholding.

DOI: 10.1504/IJMIC.2021.123362

International Journal of Modelling, Identification and Control, 2021 Vol.38 No.3/4, pp.193 - 200

Received: 01 Jul 2020
Accepted: 29 Nov 2020

Published online: 13 Jun 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article