Abstract
The assessment of ripeness is a major part of quality evaluation and depends on several factors such as soluble solid content (SSC), acidity and firmness. Most of the methods THAT measure these qualities are destructive. So it is essential to develop an efficient and nondestructive method for measuring internal attributes of fruit. The objective of this study was to investigate visible and near infrared (Vis/NIR) spectroscopy method for prediction of SSC and pH of kiwi fruit. Near infrared spectroscopy is a fast and a nondestructive analytical technique. One of the main advantages of NIR spectroscopy is that IT allows several constituents to be measured at the same time. Transmittance determinations in the 400-1000 nm range were carried out on samples which WERE separated randomly into two groups: first group for making calibration models and second one for quality predictions. Different data preprocessing and spectra treatments such as standard normal variate transformation (SNV), multiplicative scatter correction (MSC), median filter and derivative were used to eliminate noise. Then calibration models were developed by using principal component analysis (PCA) and partial least squares (PLS). Performance of different models was assessed in terms of root mean square errors of prediction (RMSEP) and correlation coefficient (r) between the predicted and measured parameter values. The correlation coefficient and root mean square errors of prediction to soluble solids content and pH were 0.93, 0.952 and 0.259˚Brix, 0.236, respectively. The results indicated the feasibility of Vis/NIR transmittance spectral analysis to predict SSC and pH of kiwi fruit in a nondestructive way.

Keywords: internal quality evaluation of fruit, kiwi fruit, Visible/NIR spectroscopy and nondestructive measurements