Pulsed Phase Thermography (PPT) transforms the pulsed thermography sequence in time domain into the frequency domain using discrete Fourier transform. This transformation is performed for each pixel. Phase data is less affected by local surface optical variations and so it is reported to detect defects located at approximately twice the depth of the temperature data. Another important advantage of PPT is that the phase angle is generally not sensitive to undesirable conditions in pulsed thermography such as non-uniform heating or surface emissivity variation.

PPT also permits quantitative analysis by estimating the depth of delaminations or flat bottom holes in composite samples. In one approach, the flaw depth is evaluated from the relationship between the relative phase angle (phase contrast between the defect and non-defective area) and length ratio (delamination depth/diffusion length). Another type of depth quantification with PPT defines a model based on thermal quadrupole theory to attain an analytical expression of phase contrast in Laplace domain where defect depth and thermal resistance appear explicitly.

In practice, quantitative PPT has been limited to the estimation of defect parameters such as depth and thermal resistance. An approach to estimate thermophysical properties, such as thermal diffusivity, through PPT had not been introduced until the quantitative PPT method discussed in this work. Here, we propose a thermal quadrupole based method that extends quantitative pulsed phase thermography. This approach estimates thermal diffusivity by solving an inversion problem based on non-linear squares estimation.

This approach is tested with pulsed thermography data acquired from a composite sample. The estimation algorithm calculates thermal diffusivity for each pixel (i,j) in the image to generate a thermal diffusivity map. We compare our results with other techniques established in the time domain. The proposed quantitative analysis with PPT provides low error estimations of thermal diffusivity. This estimation requires only the a priori knowledge of sample thickness.

Short Bio

Hernán Darío Benítez Restrepo received his undergraduate degree in Electronic Engineering and his Dr. Eng. degree in Electronic Engineering from Pontificia Universidad Javeriana Sede Cali and Universidad del Valle, in 2002 and 2008, respectively. Since February 2010 he is adjunct professor in the Laboratory of Computer Vision and Systems of Université Laval, Québec City. Since January 2008, he has been with the Department of Electronics and Computing at Pontificia Universidad Javeriana Sede Cali. In 2011, he received a Merit scholarship for short-term research from the Ministére de l’Education du Québec to pursue research on infrared vision. Dr Benitez is IEEE Senior member since 2014 and Chair of Colombia’s IEEE Signal Processing Chapter since 2012. He is a member of the scientific editorial board of the Quantitative Infrared Thermography Journal since 2014. His main research interests encompass pattern recognition, infrared vision and digital signal processing.

For more information, please consult the following website: www.hbenitez.org

Note: The seminar will be presented at 11:00 a.m. in room PLT-2508.