Classification of Wine Grape Based on Different Phytosanitary Status by Using Visible/Near Infrared Spectroscopy

The grape selection at the consignment is a delicate stage especially for big cooperative companies with thousands of members, and therefore grapes of the same variety may have different quality characteristics and health status. Currently, the quantification of diseases on wine grapes is commonly performed by a visual estimation of the proportion of area with infection symptoms in bunches and then calculating the mean severity on the sample. Depending on the grapes quality, wineries often enforce a price penalty downgrading the grapes value up to 50 % imposing a severe reduction of vine growers’ income. The application of optical techniques could minimize contentious between the members and the winery inspectors and could improve the standardization of the grape quality and therefore of the wine. The aim of this work was to investigate the applicability of vis/NIR spectroscopy for a rapid assessment of grape infection directly at the check point at the grape consignment. The experimentation was conducted in cooperation with Cantine Settesoli, Menfi (AG), Italy, using a vis/NIR device (Corona Process, Zeiss, Germany) built specially for analysis of non-homogeneous product. The system is capable to perform measurements in reflection at a variable distance of 80-600 mm between sensor and sample, in the spectral range 400-1650 nm. Spectral measurements were carried out on healthy (1235 samples) and diseased bunches (1324) from different white (Chardonnay, Grillo, Inzolia, Viognier) and red varieties (Alicante, Nero d'Avola, Syrah) for a total of 2559 spectral acquisitions. Qualitative (Principal Component Analysis, PCA) and quantitative (Partial Least Squares – Discriminant Analysis, PLS-DA) analyses were applied on grape spectra in order to test the performance of vis/NIR spectroscopy to classify healthy and infected bunches. The results obtained from PLS-DA models, in validation, gave a classification accuracy between 89.8 % and 94.0 %. Based on the confusion matrix derived from the classification analysis, the winery could apply different strategies, ranging from a more restrictive criterium with the object to preserve the wine quality, accepting the risk of erroneously downgrading healthy samples, to a less restrictive one where winery preserve the production yield, leading to the risk to accept infected samples equal to ~ 30%. Results demonstrated that the tested non-destructive system is capable to provide useful information about wine grape healthy status for a better management of the vinification process. However, further studies are needed to determine the best operating conditions for the future engineering phases to perform the measurements directly at the consignment check point station. Moreover, important factors for a success of this optical system for evaluating the infection level of grapes are the identification of the diseases and the fast quantification of the infection level, developing also quantitative predictive models.