Abstract
Different methods of data preprocessing were evaluated for the compression of Fourier transform-infrared spectral libraries by principal component analysis (PCA). The effect of noise on compressed library searches was examined. A PCA compression of an infrared library achieved an 81% reduction in size without any loss in search performance.
Get full access to this article
View all access options for this article.
References
1.
Warren
F. V.
and
Delaney
M. F.
, Appl. Spectrosc. 37 , 172 (1983 ).
2.
Lam
R. B.
Foulk
S. J.
, and
Isenhour
T. L.
, Anal. Chem. 53 , 1679 (1981 ).
3.
Hangac
G.
Wielboldt
R. C.
Lam
R. B.
, and
Isenhour
T. L.
, Appl. Spectrosc. 36 , 40 (1982 ).
4.
Rozett
R. W.
and
Petersen
E. M.
, Anal. Chem. 47 , 1301 (1975 ).
5.
Williams
S. S.
Lam
R. B.
, and
Isenhour
T. L.
, Anal. Chem. 55 , 1117 (1983 ).
6.
Malinowski
E. R.
and
Howery
D. G.
, Factor Analysis in Chemistry (Wiley-Interscience , New York , 1980 ).
7.
Gillette
P. G.
and
Koenig
J. L.
, Appl. Spectrosc. 36 , 535 (1982 ).
8.
Antoon
M. K.
D'Esposito
L.
, and
Koenig
J. L.
, Appl. Spectrosc. 33 , 351 (1979 ).
9.
Kaiser
H. F.
, Educ. Psych. Meas. 20 , 141 (1960 ).
10.
Grotch
S. L.
, Anal. Chem. 43 , 1362 (1971 ).
11.
Pouchert
C. J.
, The Aldrich Library of FT-IR Spectra (Aldrich Chemical Company , Milwaukee , 1985 ).