Three-dimensional reconstruction of subsurface defects using finite-difference modeling on pulsed thermography

Abstract

We develop a technique to analyze pulsed thermography videos in order to detect and reconstruct subsurface defects in homogeneous and layered objects. The technique is based on the analysis of the thermal response of an object to a heat pulse. This thermal response is compared to the predictions of a finite-difference model that is systematically and progressively adjusted to minimize a cost function. With this minimization process, we obtain a depth and a thickness function that allow us to determine the three-dimensional shape, size, depth, thickness, and location of internal defects. The detected defects are reliably reconstructed with graphics of easy interpretation.

Pulsed Thermal Tomography Nondestructive Examination of Additively Manufactured Reactor Materials and Components. Third Annual Progress Report

“…This is decided by positioning the IR camera at the focal length of the lens, and placing the flash lamp such that the IR field of view is not blocked, nor does IR camera block the flash. While there have been proposed a number of PIT data analysis approaches based on physics models [18,19] or machine learning [20], these methods have difficulty with detecting weak features in thermography data, which correspond to microscopic material defects. For such thermal signals, signal to noise ratio SNR < 1, with signal amplitude approaching noise equivalent temperature difference (NETD) detectability threshold.…”

Pulsed Thermal Tomography Nondestructive Examination of Additively Manufactured Reactor Materials and Components (Final Technical Report)

“…However, this method is limited to canonically shaped defects. Another model-based approach proposes the construction of computational finite-difference models to characterize defects [15]. An additional model is based on thermographic signal reconstruction (TSR).…”

Three-dimensional subsurface defect shape reconstruction and visualisation by pulsed thermography

“…Although this method can visualise the defects in the form of a 3D image, it can visualise only the depth of the defect and cannot visualise the thickness of the defect. Ramirez-Granados et al [31] proposed an approach for 3D reconstruction of subsurface defects by using a finite-difference model. Firstly, a non-defect nodal network is established with the properties and characteristics of the inspected object for detecting the inner defects.…”