Nondestructive Quality Assessment of Grape (Cv. Asgari) Based on NIR Spectroscopy

Several destructive and non-destructive methods are employed to measure the fruits and vegetables quality. Destructive methods are principally expensive and time-consuming. In the current study, some quality attributes of Vitis vinifera CV. Asgari such as soluble solids content (SSC), titratable acidity (TA), pH and total phenolics (TP) measured by near-infrared spectroscopy (NIRS). For this purpose spectroscopy was performed for 110 grape samples in the range of 900-1700 nm chemicals measures were carried out for the quality attributes of samples, and calibrated models were used to establish the correlation between the spectral data and chemical measurements. Results indicated that the best Partial least square (PLS) models had root mean square error of prediction (RMSEP) equivalent to 0.580, brix of 0.02%, 0.125 and 23.441 and correlation coefficients (rp) of 0.927, 0.806, 0.898 and 0.866 for SSC, TA, pH and TP respectively. Comparison between the mean values predicted by the best models and the mean values measured by the reference method for each attribute showed a non-significant difference between the values predicted by the best models and the measured values by reference method at the 5% level for all attributes. It was concluded that NIRs could be a fast and accurate method for non-destructive prediction of some internal quality of Asgari grape< and other attributes which are difficult to measure.