5 May 2017 Thermal NDT applying Candid Covariance-Free Incremental Principal Component Thermography (CCIPCT)
Author Affiliations +
Abstract
Thermal and infrared imagery creates considerable developments in Non-destructive Testing (NDT) area. An analysis for thermal NDT inspection is addressed applying a new technique for computation of eigen-decomposition (factor analysis) similar to Principal Component Thermography(PCT). It is referred as Candid Covariance-Free Incremental Principal Component Thermography (CCIPCT). The proposed approach uses a computational short-cut to estimate covariance matrix and Singular Value Decomposition(SVD) to obtain faster PCT results, but while the dimension of the data increases. The problem of computational cost for high-dimensional thermal image acquisition is also investigated. Three types of specimens (CFRP, plexiglass and aluminum) have been used for comparative benchmarking. Then, a clustering algorithm segments the defect at the surface of the specimens. The results conclusively indicate the promising performance and demonstrated a confirmation for the outlined properties.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Bardia Yousefi, Stefano Sfarra, Clemente Ibarra Castanedo, Xavier P. V. Maldague, "Thermal NDT applying Candid Covariance-Free Incremental Principal Component Thermography (CCIPCT)", Proc. SPIE 10214, Thermosense: Thermal Infrared Applications XXXIX, 102141I (5 May 2017); doi: 10.1117/12.2263118; https://doi.org/10.1117/12.2263118
PROCEEDINGS
7 PAGES


SHARE
Back to Top