Discrimination of flavoured olive oil based on hyperspectral imaging

Abstract

The aim of the work was to use the hyperspectral imaging as an analytical tool for the classification of flavoured olive oils, according to composition of the starting blends, crop season, flavouring methods, and flavouring agent. For this reason, in the crop season 2013–2014, three mono-cultivar extra virgin olive oils (Coratina, Peranzana, Ogliarola) were used in different proportions to produce three blends, which were successively flavoured by infusion. From each blend, 3 types of flavoured olive oils were produced: basil, chilli pepper, and garlic + chilli pepper. In the crop season 2014–2015, the study was limited to one of the blends used in the previous season and the basil, chilli pepper, and garlic + chilli pepper flavoured olive oils were produced by infusion or by malaxation. Blends and flavoured oils were submitted to hyperspecral imaging in the 400–1000 nm region. To discriminate the oil samples, ANOVA analysis was performed at each wavelength. Wavelengths in the range 400–570 nm and around 695 nm allowed the discrimination of samples.