Prediction of Chemical Contents in 'Gedong Gincu' Mango using Near Infrared Spectroscopy

The objective of this work was to predict the soluble solid content, total acid, sugar acid ratio, and crude fiber of 'Gedong Gincu' mango non destructive using Near infrared Spectroscopy. Experiments were carried out using 182 samples of 'Gedong Gincu' mango. NIR reflectance spectra measurements were performed at wavelength of 10002500 nm using NIRFlex N-500 fiber optic solid. References data were collected from laboratory measurements. Five pre-processing treatments, smoothing 3 points (sa3), normalization (n01), first derivative Savitzky-Golay 9 points (dg1), combination (n01, dg1), and the Multiplicative Scatter Correction (MSC) were used to improve the accuracy of the calibration model. Partial Least Square (PLS) method was used to calibrate NIR data through references data. The results show that the best method for prediction of soluble non solid spectra were MSC and 12 factor of PLS with calibration value of Correlation Coefficient (r), Square Error Calibration (SEC), Square Error Prediction (SEP), Ratio of standard error prediction to deviation (RPD) were 0.91, 0.25 %, 0.39 %, 2.14 respectively. Sugar acid ratio content was predictd using MSC and 12 factor of PLS calibration with values of r, SEC, SEP, RPD were 0.81, 32.08 degrees Brix/%, 38.44 degrees Brix/%, 1.45. Soluble solid content was predicted using sa3 and 15 factor of PLS calibration with values of r, SEC, SEP, RPD were 0.82, 1.04 degrees Brix, 1.28 degrees Brix, 1.52 respectively. Total acid was predicted using dg1 and 3 with the value of r, SEC, SEP, RPD were 0.74, 0.01 %, 0.12 %, 1.33 respectively. It could be concluded that the developed model could be used to predict the chemical contents of 'Gedong Gincu' mango non destructively.