PHYTOPLANKTON RECOGNITION USING PARAMETRIC DISCRIMINANTS

A comparison was made between the use of linear and quadratic discrimi nant functions for classifying phytoplankton specimens of the genera D inophysis and Ceratium by means of a general morphometric function. Th e class distributions were found to fit quadratic boundaries better th an linear boundaries. A nine species quadratic discriminant classified within 95% confidence intervals. Morphological variants not used in t he calibration were all correctly identified, although control species unknown to the model were poorly rejected. An accuracy of 99% was obt ained for separating three morphological variants of Dinophysis acumin ata. Digital filters were developed to extract the morphometric functi on directly from photomicrograph images, and present the data as an or ientation-independent feature vector. Using this feature vector, a qua dratic discriminant classified test data from 14 species of the genera Dinophysis, Ceratium and Ornithocercus with an accuracy of 83%, with 37% of the error due to two similarly shaped species of Dinophysis ove rlapping.