Premium
Discrimination and identification of morphotypes of Banksia integrifolia (Proteaceae) by an Artificial Neural Network (ANN), based on morphological and fractal parameters of leaves and flowers
Author(s) -
Pandolfi Camilla,
Messina Giuseppe,
Mugnai Sergio,
Azzarello Elisa,
Masi Elisa,
Dixon Kingsley,
Mancuso Stefano
Publication year - 2009
Publication title -
taxon
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.819
H-Index - 81
eISSN - 1996-8175
pISSN - 0040-0262
DOI - 10.1002/tax.583020
Subject(s) - proteaceae , banksia , biology , subspecies , botany , identification (biology) , ecology , woodland
Abstract An artificial neural network (back propagation neural network) based on morphological and fractal leaf and flower parameters was developed for the characterization of three Banksia integrifolia subspecies and the identification of nine unnamed morphotypes. Results indicated that the network can be effectively and successfully used to discriminate among morphotypes using simple dedicated instruments, such as a PC and an optical scanner. The new method also as a complementary approach to botanical identification, being capable of separating all the tested Banksia integrifolia accessions and of creating associations between the known subspecies and the unnamed accessions.