Artificial neural networks as a tool for plant identification: a case study on Vietnamese tea accessions

Seventeen tea accessions belonging to Chinese (Camellia sinensis), Assamic (C. sinensis var. assamica), and Shan tea (C. sinensis var. pubilimba) groups, which are either commercially planted or new promising tea germplasm, were morphologically described at Phu Tho province (Viet Nam) and assessed for their diversity. Fourteen phyllometric parameters were qualitatively and quantitatively investigated using digital image analysis. The accessions were then discriminated by a dedicated artificial neural network for univocal plant identification and a hierarchical cluster analysis was performed in order to build a dendrogram reporting the relationships among them. Results proved the diversity of investigated tea morphotypes from Phu Tho province based on a morphological screening. More, the artificial neural network was able to perform a correct identification for almost all the accessions using simple dedicated instruments.