Rapid and non-destructive prediction of mango sweetness and acidity using near infrared spectroscopy

Abstract

The study aimed on evaluating the feasibility of NIRS to predict sweetness and acidity of intact mango fruit in form of soluble solids content (SSC) and titratable acidity (TA) through calibration modeling. Diffuse reflectance spectra in NIR wavelength range of 1000 - 2500 nm were acquired for a total of 58 mango samples. PCR and PLSR were used to develop SCC or TA prediction models respectively. Multiplicative scatter correction (MSC) and standard normal variate (SNV) were applied to the spectra prior to prediction model development. The result showed that the best model for SSC prediction was achieved when PLSR is applied in combination with SNV spectra (r of calibration = 0.82) and PLSR-MSC for TA prediction (r of calibration = 0.98). These results indicated that NIRS was feasible to predict sweetness and acidity of intact mango fruit and might be considered as one of the rapid and non-destructive method of an automatic sorting and grading system based on imaging technology.