Enhanced near infrared spectral data to improve prediction accuracy in determining quality parameters of intact mango

Presented manuscript aimed to describes enhanced near infrared spectral dataset used to improve prediction performances of near infrared models in determining quality parameters of intact mango fruits. The two mentioned quality parameters are total acidity (TA) and vitamin C which corresponds to main inner attributes of fruits. Near infrared (NIR) spectra data were acquired and recorded as absorbance spectral data in wavelength range from 1000 to 2500 nm. These data were then enhanced by means of several algorithms like multiplicative scatter correction (MSC), baseline linear correction (BLC) and combination of them (MSC+BLC). Prediction models, used to determine TA and vitamin C were established using most common approach: partial least square regression (PLS) based on raw and enhanced spectral data respectively. Prediction performances can be evaluated based on prediction accuracy and robustness, by looking statistical indicators presented as coefficient of determination (R(2)) and correlation (r), root mean square error (RMSE) and residual predictive deviation (RPD). Enhanced NIR spectral dataset can be employed as a rapid, effective and non-destructive method to determine inner quality parameters of intact fruits.