ABSTRACT
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Fourier transform infrared (FTIR) techniques are well-established techniques [1-3] and have been for a long time. However, the use of Fourier transforms in the analysis of near-infrared (NIR) spectra
DBURNS: “7393_c006” — 2007/7/31 — 01:02 — page 94 — #2
NIR spectra obtained from dispersion instruments. Our [1]∗ excitement over the use of Fourier transforms in the analysis of NIR data was stimulated by work that was ongoing in the field of information processing. There had been some very exciting results in using Fourier transforms in the transmission of voice information. Electrical engineers had demonstrated that voice information could be carried in just a few lowfrequency Fourier coefficients. The work produced magnificent results. First, the voice information or signal was digitized. The Fourier transform was taken of the signals and the Fourier coefficients were transmitted down the line. However, only the first few Fourier coefficients were transmitted and, on the other end, the Fourier transform was taken again, giving intelligible voice information. Some fidelity, such as the distinction between male and female voice, was lost but the content of the information was there.
