Nondestructive Determination of Quality Management in Table Grapes using Near Infrared Spectroscopy (NIRS) Technique

Abstract

Near infrared spectroscopy (NIRS) is a potential technique for nondestructive fruit quality management. The qualities of table grapes (Vitis vinifera) as soluble solid content (SSC), pH-value, titratable acidity (TA), firmness and seedlessness are the key parameters for consumer decision. This research was focused on using NIRS for quality management in table grapes as nondestructive determination. The sample spectra were acquired by NIR spectrometer (NIRSystem 6500) on interactance mode with fiber optic probe in wavelength range of 800-1,100 nm. To develop the calibration model, the relationship between NIR spectral data and all measured properties of table grapes were studied using Partial Least Square Regression (PLSR). The results showed the best models were the positive determination in Savitzky-Golay derivative 2 of TSS, seedlessness and TA. There were coefficient of determination (R) of 0.980, 0.903 and 0.897, and standard error of calibration (SEC) of 0.430, 0.262 and 0.062, respectively. The full cross validation was analyzed, the standard error of cross validation (SECV) of 0.522, 0.570 and 0.103, and bias of -0.00272, 0.00205 and-0.00279, respectively. The R2cal and SEC of pH value of 0.632 and 0.352, and firmness of 0.661 and 0.685, respectively. Therefore, NIRS technique can be an efficiency nondestructive determination for quality management of table grapes in these key parameters.