Abstract:
Active pulse compression favourable infrared imaging techniques are promising evaluation techniques among
other thermal non-destructive testing and evaluation modalities for identification of subsurface defects in the
test samples. This is due to their merits such as high defect detection sensitivity and resolution by using relatively
low peak power heat sources and in a moderate experimentation time in comparison with widely used pulse and
modulated conventional thermal wave imaging modalities. Various data processing methods have been developed to extract the thermal features of defects from the captured thermographic data. The cross-correlation
based pulse compression method is recommended for Barker coded thermal wave imaging technique as it enhances the concentration of applied thermal energy into mainlobe and distributes the significantly less energy
into the sidelobes of cross correlation data. In this paper, the stated property of Barker coded thermal wave
imaging along with the independent component analysis approach has been utilised. The study in this paper is
focused on the application of independent component analysis approach on captured thermal response data,
entire cross correlation data and only the mainlobe of CC data. The results obtained clearly indicate that the
application of independent component analysis on cross correlation data has improved the detectability, visibility, and contrast (interms of signal-to-noise values) of the defects. Further, the results obtained with the
different methods have been compared by taking Signal to noise ratio as the figure of merit.
