Abstract: Internal quality characteristics such as soluble solids content (SSC), pH value and firmness of bruised samples of Huayou kiwifruits stored at 2 ℃ for four weeks were quantitatively evaluated by near-infrared (NIR) diffuse reflectance spectroscopy over the range of 12000－4000 cm-1 in combination with each of the three calibration methods multiple linear regression (MLR), partial least squares (PLS) regression and principal component (PCR) regression, respectively. Meanwhile, the results of PLS models based on different pretreatment methods such as the logarithms of the reflectance reciprocal D1 lg(1/R) and its first derivative D1 lg(1/R) and second derivative D2 lg(1/R) were compared. The results indicated the PLS prediction models for SSC, pH and firmness based on D1 lg(1/R) were the most accurate, exhibiting a correlation coefficient between predicted and measured values of 0.812, 0.703 and 0.919 and a root mean square error of prediction of 0.749, 0.153 and 1.700, respectively. Thus, it is feasible to use near-infrared diffuse reflectance spectroscopy to detect internal qualities of bruised kiwifruits such as SSC, pH value and firmness nondestructively.

Key words: near-infrared diffuse reflectance spectroscopy, soluble solids content, pH, firmness, kiwifruit