Evaluation of artificial neural networks and partial least squares regression for computerized interpretation of FTIR spectra

Tom Visser; Hendrik J. Luinge; John H. van der Maas

doi:10.1117/12.166787

31 January 1994 Evaluation of artificial neural networks and partial least squares regression for computerized interpretation of FTIR spectra

Tom Visser, Hendrik J. Luinge, John H. van der Maas

Author Affiliations +

Proceedings Volume 2089, 9th International Conference on Fourier Transform Spectroscopy; (1994) https://doi.org/10.1117/12.166787
Event: Fourier Transform Spectroscopy: Ninth International Conference, 1993, Calgary, Canada

Abstract

Experiments have been carried out to classify infrared spectra of pesticides with artificial neural networks and partial least squares regression into an organophosphorous and a non- organophosphorous class. The results have been compared to conclusions derived from interpretation (1) as performed by experienced spectroscopists, (2) based on literature correlation tables, and (3) by means of a knowledge based system EXSPEC. The multivariate methods applied appear to provide significantly better results than interpretation based on frequency and intensity criteria only. Classification by means of these methods approaches the results obtained from interpretation by experts. Reduction of the spectral window to a specific C-O-P equals S region does hardly affect the results. Differences between methods lie mainly in the time required for training and calibration.

Citation Download Citation

Tom Visser, Hendrik J. Luinge, and John H. van der Maas "Evaluation of artificial neural networks and partial least squares regression for computerized interpretation of FTIR spectra", Proc. SPIE 2089, 9th International Conference on Fourier Transform Spectroscopy, (31 January 1994); https://doi.org/10.1117/12.166787

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