Study on predicting sugar content and valid acidity of apples by near infrared diffuse reflectance technique

The nondestructive method for quantifying sugar content (SC) and available acid (VA) of intact apples using diffuse near infrared reflectance and optical fiber sensing techniques were explored in the present research. The standard sample sets and prediction models were established by partial least squares analysis (PLS). A total of 120 Shandong Fuji apples were tested in the wave number of 12 500-4 000 cm(-1) using Fourier transform near infrared spectroscopy. The results of the research indicated that the nondestructive quantification of SC and VA, gave a high correlation coefficient 0.970 and 0.906, a low root mean square error of prediction (RMSEP) 0.272 and 0.056 2, a low root mean square error of calibration (RMSEC) 0.261 and 0.067 7, and a small difference between RMSEP and RMSEC 0.011 and 0.011 5. It was suggested that the diffuse near infrared reflectance technique be feasible for nondestructive determination of apple sugar content in the wave number range of 10 341-5 461 cm(-1) and for available acid in the wave number range of 10 341-3 818 cm(-1).