Rootstock’s and Cover-Crops’ Influence on Grape: A NIR-Based ANN Classification Model

In this study, a multivariate analysis combined with near-infrared (NIR) spectroscopy was employed to classify intact grape berries based on the rootstock x cover crops combination. NIR spectra were collected in diffuse reflection mode using TANGO FT-NIR spectrometer (Bruker, Germany) 8 cm−1 resolution and 64 scans wave number range of 4000–10,000 cm−1. The chemometric analyses performed statistical software R version 4.2.0 (2022-04-22). Elimination uninformative variables accomplished PCA genetic algorithm (GA). discrimination performance linear discriminant (LDA) model not enhanced either PCA- or GA-based selection. A multiclass classification built an artificial neural network (ANN). best fit test data obtained GA-ANN that gave accuracy close 80% for samples belonging four classes. These results demonstrate could be used as rapid method their cover-crops