Nondestructive determination of pear internal quality indices by near-infrared spectrometry

Abstract

The objectives of the study were to establish relationships between the nondestructive near-infrared (NIR) spectral measurements and the major internal quality indices of pear ('Fengshui', Jiangxi) fruit, and to evaluate the use of NIR spectrometry in measuring the internal quality indices of pear fruit. Intact pear fruit were measured by reflectance NIR in 350-1800 nm range. In this study, Calibration models relating NIR spectra to soluble solids content (SSC), and firmness were developed based on multi-linear regression (MLR), Principal component analysis (PCA) and partial least square (PLS) regression with respect to the logarithms of the reflectance reciprocal log (1/R), its first derivative D1log (1/R)and second derivative D2log (1/R). The best combination, based on the prediction results, was MLR models with respect to D1log (1/R) at equatorial position of pear fruit. Prediction with MLR models resulted correlation coefficients (Rp) of 0.9151 and 0.8125, and root mean standard error of prediction (RMSEP) of 0.6834 and 1.3778 for SSC and firmness, respectively. The preliminary results of the built models indicated that NIR spectroscopy could provide an accurate, reliable and nondestructive method for assessing the internal quality indices of pear fruit.