Principal component analysis of thermographic data

William P. Winfree; K. Elliott Cramer; Joseph N. Zalameda; Patricia A. Howell; Eric R. Burke

doi:10.1117/12.2176285

12 May 2015 Principal component analysis of thermographic data

William P. Winfree, K. Elliott Cramer, Joseph N. Zalameda, Patricia A. Howell, Eric R. Burke

Proceedings Volume 9485, Thermosense: Thermal Infrared Applications XXXVII; 94850S (2015) https://doi.org/10.1117/12.2176285
Event: SPIE Sensing Technology + Applications, 2015, Baltimore, MD, United States

Abstract

Principal Component Analysis (PCA) has been shown effective for reducing thermographic NDE data. While a reliable technique for enhancing the visibility of defects in thermal data, PCA can be computationally intense and time consuming when applied to the large data sets typical in thermography. Additionally, PCA can experience problems when very large defects are present (defects that dominate the field-of-view), since the calculation of the eigenvectors is now governed by the presence of the defect, not the "good" material. To increase the processing speed and to minimize the negative effects of large defects, an alternative method of PCA is being pursued where a fixed set of eigenvectors, generated from an analytic model of the thermal response of the material under examination, is used to process the thermal data from composite materials. This method has been applied for characterization of flaws.

Citation Download Citation

William P. Winfree, K. Elliott Cramer, Joseph N. Zalameda, Patricia A. Howell, and Eric R. Burke "Principal component analysis of thermographic data", Proc. SPIE 9485, Thermosense: Thermal Infrared Applications XXXVII, 94850S (12 May 2015); https://doi.org/10.1117/12.2176285

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available