Feature extraction using normalized cross-correlation for pulsed eddy current thermographic images

Pulsed eddy current (PEC) thermography is an emerging method in the field of non-destructive testing and evaluation (NDT&E). Defects can be characterized by tracking the diffusion of heat in a sample through the analysis of a sequence of PEC thermographic images. This study takes advantage of the capabilities of PEC thermography to gain quantitative information about complex geometrical defects, i.e. angular defect characterization through the analysis of the surface thermal distribution. To conduct the analysis, a new approach using a normalized cross-correlation technique is proposed. The strength of the proposed approach lies in its ability to track the heat diffusion through sequential PEC thermographic images in a metallic sample. The results of the analysis are used to determine the dimensions of defects in the sample under test. These results have shown the effectiveness of the proposed technique in providing features which have good agreement with defect detection and evaluation.