Application of an artificial neural network (ANN) for the identification of grapevine genotypes

Neural networks were employed to distinguish between 15 accessions of "coloured" (fruit gives intense red colour to the wine) grapevines found in some viticultural zones of Tuscany. Our results enabled us to distinguish, with considerable certainty, between 9 accessions and to denote three pairs of synonyms. The use of neural networks opens interesting prospects for ampelography; its advantages over traditional ampelographic methods are demonstrated.