In this paper the possibility offered by a Vis-SWIR spectroscopy-based analysis is described, carried out directly in the field, to recognize post harvested olive fruit attacked by olive fruit fly (i.e. Bactrocera oleae). To reach this goal, chemometric techniques were used, that is: Principal Component Analysis (PCA) for exploratory data analysis and Partial Least Square – Discriminant Analysis (PLS-DA) for classification of attacked and un-attacked olive fruits. Itrana cultivar, at different degree of ripeness, coming from three different locations, was investigated. An ASD FieldSpec 4® Standard- Res field portable spectroradiometer working in the range 350-2500 nm was utilized. The developed classification model and the achieved results showed a promising ability to recognize attacked olive fruits, reaching Sensitivity and Specificity values in prediction of 0.939 and 0.698, respectively.
|