Automatic taxonomic categorisation of 23 species of dinoflagellates was demonstrated using field-collected specimens. These dinoflagellates have been responsible for the majority of toxic and noxious phytoplankton blooms which have occurred in the coastal waters of the European Union in recent years and make severe impact on the aquaculture industry. The performance by human 'expert' ecologists/taxonon~ists in identifying these species was compared to that achieved by 2 art~fi-cial neural network classifiers (multilayer perceptron and radial basis function networks) and 2 other statistical techniques, k-Nearest Neighbour and Quadratic Dlscnm~nant Analysis The neural network classifiers outperform the classical statistical techniques. Over extended trials, the human experts averaged 85% while the radial basis network achieved a best performance of 83%, the multilayer perceptron 66 %, k-Nearest Neighbour 60%, and the Quadratic Discriminant Analysis 56 %.
Photomicrographs of 5 species of Cymatocyl~s were digitised, binarised and edited by hand to remove large debris contaminating the images. An artificial neural network (back-propagation of error) was trained to categorise 201 of these specimens after pre-processing the data by Fourier transformation. Of the 299 trials which were carried out, 28% demonstrated better than 70% correct categorisation of the data used in the training sets. The best performing network learned to differentiate the training data set with an error rate of 11 U/;,. The same network gave an error rate of 18% when presented with previously unseen data. The results of training back-propagation of error networks are presented and the performance and limitations are discussed and compared with more classical rnorphometric and clustering techniques for the taxonomic separation of marine plankton. This automatic technique demonstrates the potential of neural network pattern classifiers for addressing the difficult taxonomic task of congeneric classification and also has wider implications for the automatic identification of field samples of marine organisms
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.