Vis/NIR spectroscopy for quick evaluation of olive fruit ripeness

Increasing consumption of olive oil and table olives has recently determined an expansion of olive tree cultivation in the world. This trend is being fostered by the recognized nutritional value of the Mediterranean diet. The aim of this work was to test a portable vis/NIR system for the analysis of ripening parameters (soluble sugar content SSC and texture) of olives before to be processed, both for fermentation process of table olives or for extraction process of olive oil. The final goal is to provide the sector with postharvest methods and sorting systems for a quick evaluation of olive fruits ripeness indices. In the present study analysis were carried out on a total of 109 olives for oil. Olives spectra registered with the optical device and values obtained with destructive analysis in laboratory were analysed. Specific statistical models were elaborated to study their correlations and to evaluate ripening indices obtained through the analysis of the visible-near infrared range (400-1000 nm). Statistical models were processed thanks to chemometric techniques to extract maximum data information. Principal component analysis (PCA) was performed on vis/NIR spectra to examine sample groupings and identify outliers, while Partial Least Square (PLS) regression algorithm was used to correlate samples spectra and physical-chemical properties. Results are encouraging: PCA showed a significant sample grouping for the different ripening stages. PLS models gave good predictive skills in validation for SSC (R2 = 0.67 and RMSECV% = 7.5%) and texture (R2 = 0.68 and RMSECV% = 8.2%).