Reduction of thermal data using neural networks

William P. Winfree; K. Elliott Cramer

doi:10.1117/12.381542

30 March 2000 Reduction of thermal data using neural networks

Proceedings Volume 4020, Thermosense XXII; (2000) https://doi.org/10.1117/12.381542
Event: AeroSense 2000, 2000, Orlando, FL, United States

Abstract

A scanned thermal line source is a rapid and efficient technique for detection of corrosion in aircraft components. Reconstruction of the back surface profile from the data obtained with this technique requires a nonlinear mapping. Neural networks are an effective method for performing nonlinear mappings of one parameter space to another. This paper discusses the application of neural networks to the reconstruction of back surface profiles from the data obtained from a thermal line scan. The neural network is found to be a very effective method of reconstructing arbitrary surface profiles. The network is trained on simulations of the thermal line scan technique. The trained network is then applied to both simulated and experimentally obtained data. The reconstructed profiles are in good agreement with independent characterizations of the profiles. Limitations of the reconstruction technique are illustrated by presenting results for several different configurations.

Citation Download Citation

William P. Winfree and K. Elliott Cramer "Reduction of thermal data using neural networks", Proc. SPIE 4020, Thermosense XXII, (30 March 2000); https://doi.org/10.1117/12.381542

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