Spectral cahbration models

First and Second Derivatives. One of the best methods for removing baseline effects is the use of derivative spectra. This method is one of the earliest methods used to attempt to correct for baseline effects in spectra solely for the purpose of creating robust cahbration models (Refs. 46, 47). The first derivative of a spectrum is simply a measure of the slope of the spectral curve at every point. The slope of the curve is not affected by baseline offsets in the spectrum, and thus the first derivative is a very effective method for removing baseline offsets (Fig. 30). The second derivative is a measure of the change in the slope of the curve. In addition to ignoring the offset, it is not affected by any hnear tilt that may exist in the data and is therefore a very effective method for removing both the baseline offset and the slope from a spectrum (Fig. 31). 

In a laboratory-based example of sensor fusion, both ISE and spectral reflectance data were obtained for 37 surface soil samples from the US states of Missouri and Illinois. Although ISE estimates of P and K were of good accuracy (r > 0.87), they were further improved (r > 0.95) by including both ISE and spectral data in the cahbration model. The authors attributed the increased accuracy to the ability of the spectral data to provide an estimate of soil texture. 

For asparagine, the best calibration model corresponds to the spectral range of 4800-4250 cm and 12 factors. The corresponding concentration correlations are presented in Figure 5. Figure 5a corresponds to the cahbration data set and Figure 5b corresponds to the prediction data set. Linear regression analysis of these data yields slopes of 0.999 0.003 and 1.039 + 0.008 and y-intercepts of 0.01 0.10 and -0.23 0.16 mM for the calibration and prediction data sets, respectively. This calibration model indicates a SEP of 0.20 mM and a mean percent error of 3.48%. 

---