Improvement of Industrial Radiography for Defect Detection of Oil and Gas Pipelines in Weld Regions by Image Processing

Subject Areas : Renewable energy

Ali Reza  Karimian ¹ (Assistant Professor/University of Isfahan)
Monir  Torabian ² (MSc/National Iranian Gas Company (NIGC), Isfahan)
Mohammad Reza  Yazdchi ³ (Assistant Professor/University of Isfahan)

Received: 2008-11-23 Accepted : 2009-10-29 Published : 2010-08-01

Keywords: image processing, weld defects, Non-destructive tests, radiographic test, NDT,

Abstract :

Industrial Radiography is one of the oldest and most usable of non-destructive methods for studying the defects inside the weld regions of pipelines. It sounds, to increase the quality of radiographic images inside the film and to decrease the weld commentary errors. It is necessary to have a system or method to improve the accuracy of recognition and detection of defects in the weld regions. In this research work, by using digital image processing methods, a new method has been proposed to improve the quality of the images on radiographic films of weld regions. The proposed method has been tested by 60 pieces of radiographic films of weld regions with different quality.  The results showed the proposed algorithm and method has the ability to detect the defects inside the weld regions with 100% precision for the films with high and normal quality and with 87% and 47% precisions for the films with low and very low qualities respectively.