Chestnut (Castanea sativa Mill.) genotype identification: An artificial neural network approach

The potential use of the artificial neural networks (ANNs) for characteriza tion and identification of seventeen chestnut (Castanea sativa Mill.) acces sions, belonging to the "marrone"-type and "chestnut"-type, was investigate d in genotypes originating from regions of Italy. Different back-propagatio n neural networks (BPNN) were built on the basis of image analysis paramete rs of the leaves, for two tasks of chestnut classification. In the first ca se a BPNN was built and trained to differentiate the 17 accessions of chest nut. In the second case a BPNN was conceived to distinguish between the "ma rrone" and "chestnut" types. BPNN produced a clear identification of all th e accessions except in the case of 'Garrone nero', 'Garrone rosso' and 'Tem puriva', which showed almost the same output diagram. Cluster analysis sepa rated the 17 chestnut genotypes into four main groups whose differences wer e related to the original sources of the genotypes and to the type of affil iation ("marrone"-type or "chestnut"-type). Artificial neural network techn ique was also able to discriminate between "marrone"-type and "chestnut"-ty pe accessions. Qualitative and quantitative rules for the image analysis pa rameters, useful for classifying chestnut accessions into these two types, were obtained. On the whole the relative importance of the leaf parameters reveals that "typical" leaves for "marrone"-type are more elongated, of a d arker colour and with a higher perimeter/area ratio than the leaves of the "chestnut"-type.