Simulation of defects in aluminium castings using CAD models of flaws and real X-ray images

In order to evaluate the sensitivity of defect inspection systems, it is convenient to examine simulated data. This gives the possibility to tune the parameters of the inspection method and to test the performance of the system in cases where the detection is known to be difficult. In this paper, an interactive environment for the simulation of defects in radioscopic images of aluminium castings is presented. The approach simulates only the flaws and not the whole radioscopic image of the object under test. A manifold surface is used to model a flaw with complex geometry, which is projected and superimposed onto real radioscopic images of a homogeneous object according to the exponential attenuation law for X-rays. The new grey value of a pixel, where the 3D flaw is projected, depends only on four parameters: a) the grey value of the original X-ray image without flaw; b) the linear absorption coefficient of the examined material; c) the maximal thickness observable in the radioscopic image; and d) the length of the intersection of the 3D flaw with the modelled X-ray beam, that is projected into the pixel. The approach allows the user the simulation of complex flaws at any position of an aluminium casting. Simulation results of flaws like blow holes and cracks on X-ray images are shown and contrasted with real digital images with real flaws.

[1]  Domingo Mery,et al.  A general approach to flaw simulation in castings by superimposing projections of 3D models onto real X-ray images , 2003 .

[2]  Nicolas Freud,et al.  A computer code to simulate X-ray imaging techniques , 2000 .

[3]  Domingo Mery,et al.  Neural network method for failure detection with skewed class distribution , 2004 .

[4]  Helmut Strecker,et al.  Automated X-Ray Inspection of Aluminum Castings , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Daniel Babot,et al.  New Developments in Virtual X-ray Imaging: Fast Simulation Using a Deterministic Approach , 2003 .

[6]  Dieter Filbert,et al.  Automated flaw detection in aluminum castings based on the tracking of potential defects in a radioscopic image sequence , 2002, IEEE Trans. Robotics Autom..

[7]  D Mery.  Flaw simulation in castings inspection by radioscopy RADIOSCOPY , 2001 .

[8]  Domingo Mery.  Explicit geometric model of a radioscopic imaging system , 2003 .

[9]  Domingo Mery.  A new algorithm for flaw simulation in castings by superimposing projections of 3D models onto X-ray images , 2001, SCCC 2001. 21st International Conference of the Chilean Computer Science Society.

[10]  F. Inanc,et al.  Scattering simulations in radiography , 1997 .

[11]  Gabriel Taubin,et al.  Cutting and Stitching: Converting Sets of Polygons to Manifold Surfaces , 2001, IEEE Trans. Vis. Comput. Graph..

[12]  Domingo Mery,et al.  A review of methods for automated recognition of casting defects , 2002 .

[13]  Greg Turk,et al.  Simplification and Repair of Polygonal Models Using Volumetric Techniques , 2003, IEEE Trans. Vis. Comput. Graph..

[14]  M. Carter.  Computer graphics: Principles and practice , 1997 .

[15]  Gerd-Rüdiger Tillack,et al.  X-ray modeling for industrial applications , 2000 .