Modelling of a high resolution digital radiographic system and development of a filtering technique based on wavelet transforms Ha ˚kan Wirdelius * , Lars Hammar Department of Mechanical Engineering, Chalmers Lindholmen University College, Box 8873, SE40272 Gteborg, Sweden Received 30 May 2003; revised 18 August 2003; accepted 8 September 2003 Abstract The first part of this paper describes a geometrical model of an iron cast object with a number of simple defects included in the volume. This model, together with a model of a specific digital X-ray system and procedure, is used to generate a simulated radiograph of the volume. This is then employed in the development of a filtering technique that extracts the grey-level information related to the existence of anomalies in the object. The wavelet-technique is applied in two different ways in order to identify the large-scale of a matrix (i.e. a radiograph). It turns out that both methods are able to remove the geometrical based information from the greyscale in the simulated radiographs. The techniques are also verified with real radiographs taken from a number of welds and iron cast objects with different kinds of defects and defect sizes. The high-resolution digital X-ray system is able to detect defects less than one percentage of the object-thickness in the range of 25 – 60 mm steel. The developed filtering technique is proven to enhance the detection of defects in objects with geometrical variations and has a potential to reduce time-consuming analysis of radiographs. q 2004 Elsevier Ltd. All rights reserved. Keywords: RT-Modelling; High resolution X-ray technique; Fibre optic scintillator; Wavelet filtering 1. Introduction According to the Swedish Nuclear Power Inspectorate’s requirements in the regulations concerning structural components in nuclear installations, in-service inspection must be performed using inspection methods that have been qualified. The qualification is based on a technical justification that provides evidence about the reliability of the proposed NDT-system and a qualification of the testing personnel working according to a well-specified procedure. A high resolution digital X-ray system was developed [1] as a collaboration between a testing company (SAQ Kontroll AB) and Linko ¨ping Institute of Technology and this system was later qualified (1998) to detect, size and characterize stress corrosion cracks in thick walled steel components in nuclear power plants. In a number of occasions NDT technicians have been qualified to detect according to their procedure but fail when it comes to characterization and sizing of the defects. This in view of the fact that utilized techniques (ultrasonic or in some case eddy-current) tends to be sensitive in their application as sizing tools when it comes to objects with complex combination of materials and geometries. The X- ray system applied in this project has a potential to be a supplement to NDT-systems that are able to detect the defects but are less successful in sizing the depth of them. The development of the technique described in this paper indicates how to make this NDT technique independent any complex combination of materials or the geometry of the object. Within a project that addresses standardization and quality control of large cast iron objects that will be exposed to fatigue stress this high resolution X-ray system (HiReX) has been used as reference system. The CCD- camera within the system generates 12 bits of information but the human eye can only separate less than 128 grey- levels, which corresponds to seven bits of information. Defect information can saturate in grey-levels caused by large variation in thickness or attenuation coefficients in the object. The defect information can though be retrieved by alteration of contrast and brightness but is a time consuming procedure. These iron cast objects often tend to be of complex geometries, which have enforced 0963-8695/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.ndteint.2003.09.001 NDT&E International 37 (2004) 73–81 www.elsevier.com/locate/ndteint * Corresponding author. Tel.: þ 46-317-722-712; fax: þ46-317-722-689. E-mail address: hakwir@chl.chalmers.se (H. Wirdelius).