A robust multi-scale gapped smoothing algorithm for baseline-free damage mapping from raw thermal images in flash thermography

Abstract

Flash thermography is a promising non-destructive testing technique for the inspection of composite components. However, non-uniform heating, measurement noise and lateral heat diffusion complicate the interpretation of thermographic measurements. In order to overcome these difficulties, a novel baseline-free processing technique called 'Multi-Scale Gapped Smoothing Algorithm' is presented. This algorithm constructs a damage map directly from the measured data, in which an (almost) zero-reference background is obtained, and where measurement noise and excitation non-uniformity are effectively suppressed. The efficiency of the proposed technique is evaluated and confirmed through synthetic data and experimental results of a carbon fiber reinforced polymer with various artificial defects.