Defect detection in the presence of geometrical artefacts

Defect detection in non-destructive testing relies on the ability to detect signals arising from defects and distinguish them from noise. Ultrasonic arrays are capable of producing multiple images through combinations of mode conversions and reflections of the wave from boundaries. While these views contain additional information about defects, they also contain an increased number of structural artefacts arising from boundary reflections, especially at large times-of-flight in complex geometries. By characterising the intensity distribution in each image within a pristine sample, defect detection can be treated as a hypothesis test. This information is combined with state-of-the-art data fusion routines, improving detection performance. Experimental work shows that this method improves performance compared to other methods of defect detection over the same scan area. Equivalent performance is found to methods that feature additional prior knowledge such as the defect type and location, thus providing a viable method of defect detection in the presence of large structural artefacts.