Ultrasonic imaging of complex specimens by processing multiple incident angles in full-angle synthetic aperture focusing technique

In the evaluation of large-scale metallic specimens, X-ray CT suffers from limited penetration, which results in artifacts in the reconstructed image. Data fusion of information obtained by different modalities allows correction of those artifacts. In this contribution, an approach is presented to provide complementary data of the inner pattern of the specimen by ultrasonic testing in immersion mode. To process an ultrasonic imaging full-angle synthetic aperture focusing technique, data are acquired along the a priori known contour of the specimen. Substantial discrepancies in speed of sound between the couplant and the material of the specimen lead to refraction effects which are corrected by a virtual source element method. Furthermore, several incident angles at each virtual source are utilized to achieve an enhanced detectability of inner structural edges. However, arising reverberations limit image quality and must be suppressed by predictive deconvolution. Additionally, a subspace analysis and projection method is utilized to remove echoes of the a priori known surface in the reconstructed image which potentially mask information of near-surface structures. In comparison with exclusively perpendicular insonification, resulting images show a significant enhanced possibility of detection for inner structural edges even in adverse orientations for ultrasonic imaging. Furthermore, surface echoes and reverberations are suppressed by the proposed filter methods in a reliable way.