A convenient algorithm for optimizing wavelength selection in multiple linear regression (MLR) has been developed. MOP (MLP Optimization Program) has been developed to test all possible MLR calibration models in a given spectral range and finally find an optimal MLR model with external validation capability. MOP generates all calibration models from all possible combinations of wavelength, and simultaneously calculates SEC (Standard Error of Calibration) and SEV (Standard Error of Validation) by predicting samples in a validation data set. Finally, with determined SEC and SEV, it calculates another parameter called SAD (Sum of SEC, SEV, and Absolute Difference between SEC and SEV: sum(SEC+SEV+Abs(SEC-SEV)). SAD is an useful parameter to find an optimal calibration model without over-fitting by simultaneously evaluating SEC, SEV, and difference of error between calibration and validation. The calibration model corresponding to the smallest SAD value is chosen as an optimum because the errors in both calibration and validation are minimal as well as similar in scale. To evaluate the capability of MOP, the determination of benzene content in unleaded gasoline has been examined. MOP successfully found the optimal calibration model and showed the better calibration and independent prediction performance compared to conventional MLR calibration.

A variety of standardization methods between two near-infrared (NIR) spectrometers were investigated for the prediction of five constituents in trans-alkylation process. Spectra were collected by two different instruments (one is regarded as mater instrument, other on as slave instrument). Three well-known standardization methods of direct standardization (DS), piecewise direct standardization (PDS) and slope/bias correction of response variable were applied to trans-alkylation samples on the slave instrument. We have examined for a set of reliable standardization samples using smaller number of transfer samples in order to increase efficiency of standardization.

The application of near infrared reflectance spectroscopy to determine the cultivation years of ginseng is reported. The objective of this study is to develop a nondestructive and accurate method using near infrared (NIR) reflectance spectroscopy and pattern recognition techniques. Four 5, and 6-year-old ginseng samples are studied. Though there are no serious difference in the NIR spectra according to cultivation years, the spectra were moderately differentiated based on cultivation years using pattern recognition method based on principal component analysis. It is shown that there are qualitative differences according to cultivation years. In order to develop classification rules, two pattern recognition techniques such as discriminant partial least squares (PLS2) method and soft independent modeling of class analogies (SIMCA) have been carried out. As a result, near infrared reflectance spectroscopy using pattern recognition is shown to have significant potential as a rapid and nondestructive method for the determination of cultivation years of ginseng.

The objectives of this study were to examine the ability to predict soluble solid and firmness in intact apple based on the visible/near-infrared spectroscopic technique. Two cultivars of apples, Delicious and Gala, were handled, tested and analyzed. Reflectance spectra, Magness-Taylor (MT) Firmness, and soluble solids in apples were measured sequentially. Maximum and minimum diameters, height, and weight of apples were recorded before the MT firmness tests. Apple samples were divided in to a calibration set and a validation set. The method of partial least squares (PLS) analysis was used. a unique set of PLS loading vectors (factors) was development for soluble solid and firmness. The PLS model showed good relationship between predicted and measured soluble solids in intact apples in the wavelength range of 860∼1078 nm. However, the PLS analysis was not good enough to predict the apple firmness.

In this study, we investigated the feasibility of non-destructive determination of internal quality factors of Fuji apple fruit using near infrared(NIR) reflectance spectroscopy and developed the calibration models. As the reference methods, refractometer, titration and texture analyzer for sugar content, acidity and firmness were used, respectively. Samples were scanned from 1100∼2500nm with InfraAlyzer 500C spectrometer and SESAME software was used for data analysis. A multiple linear regression(MLR) analysis was performed to develop the calibrations. The correlation coefficient(R) and standard error of prediction(SEP) were as follows; 0.91, 0.41$^{\circ}$Brix for sugar content, 0.90, 0.04% for acidity and 0.84, 0.094 kg for firmness, respectively. This study shows that NIR spectroscopy can be used to evaluate the sugar content acidity and firmness of apple fruit with acceptable accuracy.

Near-infrared reflectance spectroscopy(NIRS) was used to determine the humic acids in soil samples from the fields of different crops and land-use over Youngnam and Honam regions in Korea. An InfraAlyzer 500 scanning spectrophotometer was obtained near infrared relectance spectra of soil at 2-nm intervals from 1100 to 2500nm. Multiple linear regression(MLR) or partial least square regression (PLSR) was used to evaluate a NIRS method for the rapid and nondestructive determination of humic acid, fulvic acid and its total contents in soils. The raw spectral data(log 1/R) can be used for estimating humic acid, fulvic acid and its total contents in soil by MLR procedure between the content of a given constituent and the spectral response of several bands. In which the predicted results for fulvic acid is the best in the constituents. The new spectral data are converted from the raw spectra by PLSR method such as the first derivative of each spectrum can also be used to predict humic acid and fulvic acid of the soil samples. A low SEC, SEP and a high coefficient of correlation in the calibration and validation stages enable selection of the best manipulation. But a simple calibration and prediction method for determining humic acid and fulvic acid should be selected under similar accuracy and precision of prediction. NIRS technique may be an effective method for rapid and nondestructive determination for humic acid, fulvic acid and its total contents in soils.

The purpose of this research was to develop a the reflection technique with near infrared (NIR) radiation for estimating soil components. NIR reflectance was scanned at 2nm intervals from 1100 to 2500nm with an InfraAlyzer 500 (Bran & Luebbe Co.). Over 400 soil sample from fields of different crops and land-use over Youngnam and Honam regions were used to obtain mean diffuse reflection of the soil for the calibration and validation of the calibration set in estimating moisture, organic matter (OM) and total nitrogen (T-N) of the soils. Multiple linear regression (MLR) was used to evaluate the correlation of NIR spectroscopy method. Reflection pattern of NIR spectra for finely sized sample (<0.5mm) and coarsely sized soil(<2mm) did not show much difference. The results showed that NIR spectroscopy and coarsely sized soil (<2mm) did not show much difference. The results showed that NIR spectroscopy could be used as a routine soil testing method in estimating OM, moisture, T-N in soil samples simultaneously.