| ABSTRACT: | AUTOMATIC ALGORITHM FOR QUANTITATIVE PULSED PHASE THERMOGRAPHY
CALCULATIONS
C. Ibarra-Castanedo, D. Gonz‡lez, X. Maldague
Université Laval, Quebec, Quebec, Canada
Pulsed Phase Thermography (PPT) is a NDT technique combining interesting features from Pulse
Thermograhy (PT) and Lock-In Thermography (LT), i.e. PPT is a rapid and easy to deploy technique (as
PT) providing phase delay images (as LT) after processing. It is well known that phase is less affected by
non-uniform heating, surface emissivity variations and non-planar surfaces. As a result, PPT has been
successfully applied for defect detection purposes. In contrast, a great deal of work has been carried out in
order to evaluate the potential of PPT for quantitative applications using Statistical methods, Neural
Networks or Wavelet Transform. However, calibration requirement and lengthy computation subroutines,
precludes its use on most NDT applications. A new inversion technique, based on phase delay data, has been
recently proposed by the authors. Quantification is carried out by correlating defect depth with its
corresponding blind frequency, i.e. the frequency at which the defect becomes visible for the first time on
the phase spectra. Estimation of the blind frequency is performed however using phase contrast calculations,
which means that a non-defective zone on the sample surface is needed on the computations. On the other
hand, as will be stressed on this paper, there exists an “optimal” sampling frequency at which the phase for a
given defect depth can be calculated without aliasing. This frequency is in fact equivalent to the blind
frequency, the only two requirements are that (1) the temporal truncation window is large enough to hold all
the important thermal information, and (2) the sampling theorem is respected for all defects. We propose
here an algorithm for the automatic location of the optimal sampling frequency using phase profiles instead
of phase contrast profiles, eliminating in this way the need of a sound area.
|